Chapter10_Unsupervised_Learning.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Chapter 10 - Unsupervised Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "[10.2 Principal Components Analysis](#10.2-Principal-Components-Analysis)\n",
    "> [10.2.1 What Are Principal Components?](#10.2.1-What-Are-Principal-Components?)<br>\n",
    "> [10.2.3 More on PCA](#10.2.3-More-on-PCA)\n",
    "\n",
    "[10.4 Lab 1: Principal Components Analysis](#10.4-Lab-1:-Principal-Components-Analysis)\n",
    "\n",
    "[10.5 Lab 2: Clustering](#10.5-Lab-2:-Clustering)\n",
    "> [10.5.1 K-Means Clustering](#10.5.1-K-Means-Clustering)<br>\n",
    "> [10.5.2 Hierarchical Clustering](#10.5.2-Hierarchical-Clustering)\n",
    "\n",
    "[10.6 Lab 3: NCI60 Data Example](#10.6-Lab-3:-NCI60-Data-Example)\n",
    "> [10.6.1 PCA on the NCI60 Data](#10.6.1-PCA-on-the-NCI60-Data)<br>\n",
    "> [10.6.2 Clustering the Observations of the NCI60 Data](#10.6.2-Clustering-the-Observations-of-the-NCI60-Data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib as mpl\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.pipeline import make_pipeline\n",
    "\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.cluster import KMeans, AgglomerativeClustering\n",
    "from scipy.cluster.hierarchy import dendrogram\n",
    "from sklearn.metrics.pairwise import pairwise_distances\n",
    "\n",
    "%matplotlib inline\n",
    "plt.style.use('seaborn-white')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10.2 Principal Components Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 10.2.1 What Are Principal Components?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Murder</th>\n",
       "      <th>Assault</th>\n",
       "      <th>UrbanPop</th>\n",
       "      <th>Rape</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Alabama</th>\n",
       "      <td>13.2</td>\n",
       "      <td>236</td>\n",
       "      <td>58</td>\n",
       "      <td>21.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Alaska</th>\n",
       "      <td>10.0</td>\n",
       "      <td>263</td>\n",
       "      <td>48</td>\n",
       "      <td>44.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Arizona</th>\n",
       "      <td>8.1</td>\n",
       "      <td>294</td>\n",
       "      <td>80</td>\n",
       "      <td>31.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Arkansas</th>\n",
       "      <td>8.8</td>\n",
       "      <td>190</td>\n",
       "      <td>50</td>\n",
       "      <td>19.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>California</th>\n",
       "      <td>9.0</td>\n",
       "      <td>276</td>\n",
       "      <td>91</td>\n",
       "      <td>40.6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Murder  Assault  UrbanPop  Rape\n",
       "Alabama       13.2      236        58  21.2\n",
       "Alaska        10.0      263        48  44.5\n",
       "Arizona        8.1      294        80  31.0\n",
       "Arkansas       8.8      190        50  19.5\n",
       "California     9.0      276        91  40.6"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_heart = pd.read_csv('Data/USArrests.csv', index_col=0).dropna()\n",
    "df_heart.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = df_heart.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaled_pca = make_pipeline(StandardScaler(), PCA(n_components=2, whiten=False))\n",
    "pca = scaled_pca.named_steps['pca']\n",
    "pcaX = scaled_pca.fit_transform(X)\n",
    "pcaX = pcaX[:,:2]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### FIGURE 10.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# the second pca component is inverted so the plot matches the book\n",
    "# the original features as a function of the principal components have been scaled by 2 so they can be seen easier\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(8,8))\n",
    "\n",
    "ax.scatter(pcaX[:, 0], pcaX[:, 1], s=0)\n",
    "ax.set_xlabel('1st PC')\n",
    "ax.set_ylabel('2nd PC')\n",
    "\n",
    "for i, txt in enumerate(df_heart.index):\n",
    "    ax.annotate(txt, (pcaX[i, 0], -pcaX[i, 1]), horizontalalignment='center', verticalalignment='center', color='b')\n",
    "    \n",
    "components = pca.components_\n",
    "for i, col in enumerate(df_heart.columns.tolist()):\n",
    "    ax.annotate('', xy=(2*components[0, i], -2*components[1, i]), xytext=(0, 0), arrowprops=dict(arrowstyle=\"->\", ec=\"orange\"))\n",
    "    ax.text(2*components[0, i], -2*components[1, i], col, size=15, color='orange')\n",
    "    \n",
    "ax.set_ylim(ax.get_xlim());"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 10.2.3 More on PCA"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Pipeline(memory=None,\n",
       "     steps=[('standardscaler', StandardScaler(copy=True, with_mean=True, with_std=True)), ('pca', PCA(copy=True, iterated_power='auto', n_components=None, random_state=None,\n",
       "  svd_solver='auto', tol=0.0, whiten=False))])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "scaled_pca = make_pipeline(StandardScaler(), PCA())\n",
    "pca = scaled_pca.named_steps['pca']\n",
    "scaled_pca.fit(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### FIGURE 10.4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10,4))\n",
    "\n",
    "ax1.plot(range(0, pca.n_components_), pca.explained_variance_ratio_, '.-')\n",
    "ax1.set_ylabel('Prop. Variance Explained')\n",
    "\n",
    "ax2.plot(range(0, pca.n_components_), np.cumsum(pca.explained_variance_ratio_), '.-')\n",
    "ax2.set_ylabel('Cumulative Prop. Variance Explained')\n",
    "\n",
    "for ax in (ax1, ax2):\n",
    "    ax.set_ylim(bottom=-0.05, top=1.05)\n",
    "    ax.get_xaxis().set_major_locator(mpl.ticker.MaxNLocator(integer=True))\n",
    "    ax.set_xlabel('Principal Component')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10.3 Clustering Methods"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10.4 Lab 1: Principal Components Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Already done in [10.2 Principal Components Analysis](#10.2-Principal-Components-Analysis)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10.5 Lab 2: Clustering"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 10.5.1 K-Means Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# generate random data\n",
    "X = np.random.normal(size=(50,2))\n",
    "X[:25, 0] += 3\n",
    "X[:25, 1] -= 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
       "    n_clusters=2, n_init=20, n_jobs=1, precompute_distances='auto',\n",
       "    random_state=None, tol=0.0001, verbose=0)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "kmeans = KMeans(n_clusters=2, n_init=20)\n",
    "kmeans.fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1fd11882208>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWwAAAD2CAYAAADyFb0oAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3Xd4VFX+x/H3nT6pQIAEQm+HIqCigCsoiKhYELDj6ooo6m91raugrl1xLWtvK2AD115W14JgRURRlCJw6T0JIUBIMpl67++PASTJBEKY5M4k39fz+DzmTjLzeUjyzZlzz/kezTRNhBBCJD6b1QGEEELUjBRsIYRIElKwhRAiSUjBFkKIJCEFWwghkoQUbCGESBKO2nyRUsoJTAM6AG7gPl3X/xvHXEIIISqpVcEG/gwU6bp+kVIqC/gV2FuwlVJu4GggD4gcckohhGgc7EArYL6u64HKD9a2YL8NvLPPx+FKjx8NfFfL5xZCiMZuMDCn8sVaFWxd10sBlFLpRAv37ZU+JQ9gxowZ5OTk1OYlhBCi0cnPz+fCCy+E3TW0stqOsFFKtQXeB57Vdf31Sg9HAHJycmjTpk1tX0IIIRqrmFPJtb3pmA3MBK7WdX32oaQSQghRM7UdYd8KNAX+oZT6x+5rI3RdL49PLCGEEJXVdg77WuDaOGcRQgixH7Wew64rO4qDLPq9mLRUB4cf1gS7XbM6khBCJISEKtivvrWel99Yj8MR3YDp9dh57N4+dGqfanEyIYSwXsJsTV+waAevvrWBYMjEVx7BVx6haEeQm+5ajGHIIQtCCJEwBfv9T7bgDxhVrpeWhVm6osSCREIIkVgSpmCXlFbeLBmlaeDzxX5MCCEak4Qp2EMHtcDjrhonHDE5rEemBYmEECKxJEzBHjEsh47tUvF4opFsNnC7bVw3oQspXrvF6YQQwnoJs0rE5bTxzD8PZ/Z3W/n2h200zXRy5imtUV3SrY4mhBAJIWEKNkSL9ogTchhxgjSMEkKIyhJmSkQIIcT+ScEWQogkIQVbCCGShBRsIYRIElKwhRAiSUjBFkKIJCEFuxEyDJMNm3wUFPqtjiKEOAgJtQ5b1L2fF+7g3keXU+YLYxjQsV0K903qRatsj9XRhBAHIAW7gYpETF5/byPvfryZMl+Eww/L5LxRbZh475IKXRFXri3lmlt/460XB2CzyWERQiQyKdgN1D+f1pn9XSGB3cV53i/b+fm3HRhmxd7ihgG7SsL8ungn/fo2tSKqEKKGpGA3QNuKAsz6ZivB0B/F2TSjnQ/NGGdBmCZs2x6sx4RCiNqQm44N0NqNPlyuqt9a04z2F68sYpj0Uhn1kEwIcShkhF2N9Rt9/PjrdlK8do47pjkZaU6rI9VYbo6nwuh6D7stek5mMGTsfdzjtjF0UAvatPbWd0whxEE6pIKtlBoA/FPX9SHxiWM90zR5aspqPvgsD0wTm13j8RdWMfm2Xhx9RDOr49VI6xwv/fo04ZeFOwmG/rjB6HTaeOy+Psz5sYgv5xTi9dgZc2prTj1RuiMKkQxqXbCVUjcDFwFl8YtjvV8W7eS/n+cRDO4udLtHordNXspH0/+EO8ZUQyK695aePP7vVXz+dQGRiEm73BT+/tdu9OiaQY+uGVz+545WRxRCHKRDGWGvBsYAr8UpS0L4ZFZ+zMOANS16svsxR2VZkOrgeTx2Jv5NcdNfuxEOGXg8NTu1Z+XaUh57fiVLlu3C7bYz8uQcJlzcKWn+UAnRkNX6t1DX9XeBUByzJIRIJMYyCgBzP48lMIddq3Gxzt/q56+3/MaipbswTCj3R3j/0zzufGhpHacUQtSEDJsqGT6kJV5PjMOADZMj+zTsdcpv/3dzhTlvgGDQ4KcFO9icX25RKiHEHlKwKzn26CwGDWiOx21DAxwODbfLxm3XqQZ/GPCKNSWEw1XfRTidGhs2+SxIJITYlyzrq0TTNO64sTuLl+1i7s9FpKU4OPG4luS0bPi9NlTndBYv21WlaIdCJu3apFiUSgixxyEVbF3X1wED4xMlcWiaRp+emfTpmWl1lHp19hm5/PfzPMLhyN5rLpeNAUc2JTfH+nXas74t4MXX1lGwLUDrHA//d0knBg1oDsCCxTt5/uU1rNvoo3WOh8v/3JFj+yfHDWIhakqmRMReOS09PPvPw+nbKxObDVK8dkaPaMXdN/e0Ohqfzs5n8pMr2JzvJxw22bCpnDsfWsacn7bx88Id/P3uxSxdUYKvPMKqtWXc8dBSZn271erYQsSVTImICrp0TOOZBw+3OkYVL7y6dm8jqz0CQYPnX16L02mr+ljA4Jlpaxg2uAVarP34QiQhKdgi4YXDBkU7Yjen2pRXHrM/CkDR9gChsInLKQVbNAwyJSISnt2u0SQjdi+X7BZuWmS5Yz6WmurA6ZBiLRoOKdgNmGmafPDpFs66dB4njPmOCTcuYNHSYqtjHTRN07h0bAc87oo/rh63jQkXdeTSC9rHfOzP57ST6RDRoMiUSAP26lsbeO3tDXu32i9dUcINdyzi6cmH071rusXpDs6oEa0AmPb6OnbuCtG8mYsrLu7IsMEtAfCVR5gyfR3l/ggul42xY9oydnQbKyMLEXdSsBuoYMjgtXc2VOmLEggaTJmxjkfu6m1RstrRNI3Rp7Zm9KmtCUdMHPaKI+cxp+UyakRrSsvCpKY4sNtlZC0aHinYDdS2ogBUc7rMqrWl9R8ojioX6z1sNo2M9OTpWy7EwZI57AaqWRNXzOPAANrmWr8JRghx8KRgN1Aej53Rp7WucjPO7bZx6QUdrAklhDgkCT8lklfgZ93GMtq09tK2tfSzOBhX/aUTHreNtz7cjN8fIbulh2sv78wRvZtYHU0IUQsJW7BDIYN7Hl3G9z9tx+nUCIdN+h6WyQOTetW4v3NjZ7drXHZhR8aP7bB7A4m8oWpoAgXbCO0oJqVLe2yOhP11FnGSsL/BL7+xnrnztxMMGZT5IgSCBr8tLuapqautjpZ0NE2TYt3ABIt28OMp4/iy81DmHHM2s3L/xJa3P7U6lqhjCftb/MGnWwgEKzXTDxl8+mUBhpF8J78IEU8/j/4/ir79CSMQJFLqI7S9mEWXTWTn/EVWRxN1KGELts8fiXk9FDKkYItGrXTFWop/W4oZCle4HikPsObxlyxKJepDwhbsI3s3idnUR3VJx+FI2NhC1LlAfiE2Z4z5atOkfN3m+g8k6k3CVr6/XdaFFK99b/Meh0PD67Fx0/91tTiZENbK6K0wglW7F2puF1nD/mRBIlFfEva2cvu2Kcx49mje+WgzS1eW0KVDKueMbEOr7IZ1VFdegZ9vfijEMGDQgCza5crSRbF/zqaZdLrpctY+No1IWfRwZM3pwJmZTsdrLrY4nahLCVuwAZpnubnykk5Wx6gz73+ymaemrsE0TExg6ox1XHJeOy46t73V0USC63bHNWT06sqax14iuG07LU45ji4Tr8TdopnV0UQdSuiC3ZBt3RbgqSlrCIb+WAkTxuTlNzcwaGBzOrZLtSzb7O+2MuPdjezYGaRf36aMH9uhwb2zSXaaptHq7BG0OnuE1VFEPUrYOeyG7rt520CrutolHDb4Zm6hBYmiXn1rPZOf0FmxupTCoiAzvy5g3LW/UFDotyyTECJKCnYCqq5pU13zlUd45c2KLVkNA/yBCDPe3WhNKCHEXrWeElFK2YBngb5AALhM1/VV8QrW0A0akMUzL62hcg9Uh8PGkGNbWJJpwyYfDodGoNIChHDY5NfFOy3JJIT4w6GMsEcBHl3XjwEmAo/GJ1LjkN3CwzXjO+Fy2XA6NBwODbfLxl/Oa2/Z/HXzZi5CISPmYzKHLYT1DuWm4yDgMwBd1+cppY6KT6TGY/SpuQzsl8XXcwsxDJPBA5tbuqyveZabfn2b8vPCHYRCf4z8PW4bF57VzrJcQoioQxlhZwD7nugaUUrJqpOD1CrbwwWj23LhWe0SYg323Tf35Nijs3A6NTxuG5npDm65pht9e2VaHU2IRu9QCuwuYN+TXG26roer+2SRHFK8du6b1IuS0jC7SkPktPBUOR9x67YAX39fSChsMKh/c9q3tf4PjRCNwaEU7O+BM4C3lFIDgcXxiSQSQXqag/S0qj8en8zK55HnVoJpYhgw7fX1XDCmDZdd2NGClEI0LocyJfI+4FdKzQUeA66PTySRqHbsDPLIcysJBg2CIZNwxCQQNPjPe5vQV5VYHU+IBq/WI2xd1w3gyjhmEQnu+/lF2GJ0UAyGDL6cU4jqkl71QRGTEQyy9dNvCeQX0uzYfqQf1s3qSCIJyE1CUXP72dAjLcprrnT5an4YdhGRcj9mONr3PXvkMI545WE0uxx/J6onOx1FjR3bPwsjxjJtl8vGsMHWbPZJNqZp8vNZfyVYuJ1ISRlGuR+j3M/Wj75k06vvWx0vboxQiIKPv2TdczMo/mWJ1XEaDBlhixpr2sTFjVd14dHnV4FpEjFMHHYb553Zhu4yHVIjvlXr8W/Kq9J/IOIrZ/2/36DtuLMtShY/vrUbmTv0QsK7SjHDETRNo9lxR3PUe89iczqtjpfUpGCLg3La8FYcdXhTvvq+kFDIZPDALDq0ta6zYLIxAkGwxX5jawQC9Zymbiy48AYCeYXs+3as6JufWPvkK3S+8TILkyU/KdjioGW38HD+qLZWx0hKaT274EhLIVLqq3Dd5vXQ+vwzLEoVP4GCbZQsWk7luTOj3M/GqW9LwT5EMoctRD3SbDaOmP4v7KlebG4XAPa0FNK6d6Lj1RdZnO7QGaEwMQ9jBYxgqJ7TNDwywhainmUdP4AhSz9n4yvv4d9cQPOhA8keOaxBzO96crPxtm1F2cp1Fa7b3C5an3uqNaEaECnYQljA0zqbrpOusjrGIdv29TyW/f1BSpasxNUyiy4Tr6DvKw/z08mXYITDGOUB7GkpeNu0ovNE2bZxqKRgi4SzfqOPdZt8dGiTIn1KEtiOH35l/plXYPiipxEFthSwfOLDdJ54BUP0WWx67X3K122m2bH9yBk9HJvLZXHi5CcFO8HsKg2xrShIq2wPXk/j2kQRCESY9MDvLFxSjMOhEQ6b9O2VyeTbeuF2J/6/hREOs/Hld9k45W3MSIQ2F4+m3YTzsbsbZqHS73x8b7HeI+IrZ/VD/6bTDePpfMN4i5I1XFKwE0QoZPDQMyuY/e1WHA4bkYjJBWPaMn5se7RqbuI0NM++vIbfFu8kGDL3nnrz25KdPPvyGq6/oqu14Q7ANE0WnPc3ts2aS8RXDoC+Yg35733OwNmvoVWzlC+ZlSyt5oApwyRYsA1vu9b1G6gRaHg/RUnq6Wmr+fK7QoIhE195hEDQ4I33N/LRzDyro9WbT2YVEAxV3FASDJl8MivfokQ1Vzx/cYViDRDx+Sn+bSmFM7+zMFndSevaIfYDGrhaZtVrlsZCCnYCCIcNPp6ZTyBYce2qP2Aw/Z3Gc/htIBip5nrsY8sSyfbvf8EIV20HHyn1sf27n6v9Ov/mAnzrN2NadfLyIeh297XYvBWPjrOneOlwzcXYPe4aPYcRDpP/4SxWPfRvCj6aHfPfUPxBpkTiZNa3BbzwyjryC/1kt3Az4aKOnDQku0Zf6w8YRCKxf2F3Fjeetat9e2by2+/FFXZta1r0eqJzZ2dhczmJVFprbPO6cedU7bNSumItCy64lrIV60CLrho5Yvq/aHJU73pKfOiyjutPvzef5Pcb7se3aj3Oppl0uukyOt9Us80xga1FzB18HoHC7UR8fuxeD+6c5vzp2zdwt2hWx+mTk4yw42DWtwVMfnIFeVv9mCbkbw3w0NMr+Pyrghp9fWqKnaxmsW9M9VKNp0fHDVd1JcVrx+mMztk7nRopXjvXX5nY89cAOaOGY3NWHf9oNju5F5xe4VokEOSHIWMpWbwCwx/AKA/gW72BH0/6C8GiHfUVOS5ajjieoctmcqp/KSdt/YkuN0+o8Xz979feQ/mGPCIlZRCJECkto3z9ZpbecH8dp05eUrDj4IVX1hEIVJ3O+Pdra2v09Zqmcf2VXXC7//h22Gzg9dj4v0s7xzVrIuvYLpUZzx3N2DFtGdivGWPHtGXGc0fTqX3i9yqxp3gZ+MVreDu0wZ7qxZ6Wgrt1S/r/bwqu5hVHi1s//pKIP1ClAZQRjrD5Px/VZ+y4Odi2sKZpkv/hbMxKUyBmKEz++zPjGa1BkSmROMgv9Me8XlAYwDTNGq3yGNS/OY/f24dX3tzAxi0+enRN55Lz2ze6xkrNm7m5/M/JedxYRt/uDF0xi9LlazDDYdJ7dY052vRvLsCMsU3bKPdTvn5LfURNEMk3b281KdhxkN3CTf7Wqp3Wslu4D2pJXu8emTxyV/LMYYqqNE0jvcf+3xU16d8XzWGHSj8y9rQUmh3bb79fGyn3s/V/XxEs2knW0IGkdUvOP26aptHy1CEU/O8rCP9xs1lzOMg+4wQLkyU2mRKJgwkXdawwnQHgdtu4/M8drAkkElqTAX1p+qd+FVZY2DwuUju3p+XpQ6v9uuJfljCr3WAWXn4bS//+IN/1O5PFV9+ZlCtMAA578g48OS2wp0XfRdrTUvC0bknPx263OFnikhF2HJw0JBvThH+/tpaCwgAtm7uZcFEHTjkhJ66v4yuP8O0P29hRHOSIw5rQvWvjuSHZkGiaxtEfPMe6Z6azYdrbmKEwuReOpNMN47E5Yv9KmpEI80ddSXjnrgrXN0//kBYnDiJn1PD6iB5XntbZDFn+Bfnvz6R0+WrSe3Qhe9TwBrszNB6kYMfJyUOzOXlodo3nrA/W8pUlXHf7QiIGhMIGDrvGwH7NuPvmntjtDW8nZLk/wsdf5DHnxyKaNnFx1mmt6d0j8Zf31ZTN5aLT9ZfS6fpLa/T5O+cvJlzmq3I9UlbOhilvJmXBBrC7XeSef/qBP1EAUrDjri6KtWGYTLr/d0p9f8z1hcMm837ZzudfFXDqifEdyVvNVx7h8hsXkF/gJxA00DSYM28bV4/vzKgRjXO7sxEKVfuzZezZxy8avEOaw1ZKjVZKvR6vMCK21evLKC2rugPMHzAa5Nb1j2bm7S3WEF395g8YPD11Nb7y2LshG7om/fvGvG5P8ZI7dmQ9pxFWqXXBVko9AUw+lOcQNWNETKhm4G4YyXnDaX++mVsYczu63a6xbMWuGF/R8NndLg5/5WFsXg+aK3rQgT0thSb9+5D75zMtTifqy6FMicwFPgCuiFMWUY0uHdPwuG2UVxpdetw2Rgyr2fb3ZNIkI/bJKxHDJD2t8c7iZZ9+Ascv/oRN0z8guLWIFicfR8uTBx/0phWRvA7406+UGg9cX+nyOF3X31RKDamTVKICu13j3lt68ve7F2MY0WZIXo+NniqD04e3sjpe3J19Ri4//boD/z67RzUNWmS56dopzcJk1ktpn0u32/5qdQxhkQMWbF3XpwJT6yGL2I/DD2vCW1MGMOubrWzfGeTIPk3p16cJNlvDWyFyZJ+mjL+wAy9OX4fToWEY0Kypk0fu6t1oeoMLEUvjfX+ZhJpmujhnZBurY9RKIGgw+9utzP9tB9kt3Jx5SmtaZXuq/fwLRrfl9OGtWLpiFxlpDrp3TZdiLRo9Kdj1qLQszFNTVjP7u62EIyYDjmzG9Vd0Iadl9YWrISjzhZlw068UFPrx+w0cDo13PtrM5Nt6cfQR1bfRTE9zMOBIabMpxB6HtMJD1/WvdV0/P15hGjLTNLnm1oXM/LoAf8AgHDb5YX4Rl9+4AJ+vYTdtf/ODTeTlR4s1RNeQ+wMG9/5redKscgkVl1Dw8Zds++qHvU32TcNg9WPTmNX+OD7N6Mu84RdT/Nsyi5OKhkxG2PVk4e/FbNpSTij8R4EyTCgvjzDzmwJGjci1MF3dmj2nkGCo6jK9cn+E9Zt8dGyX2B0J1095k6XX34/N5cQ0TeweN/0/nsKm6R+wcepbRHYfRFv09Y/8MGQsg356L2mbMonEJmuo68naDb6Yo0l/wGDlmjILEtUfjzv2j1nEAE+Cn4Ze/Nsylt7wAIY/QHhXKZGSMoKF25l3yjg2vPjm3mK9h+EPsPqfL1iUVjR0UrDrSfu2KcQ6iMPjttGlY2KPMA/VWaflVinamgbtc737vfGYCDZMeRMzWHXrtxkIEOsbakYi7Px5cX1EE42QFOx9+Moj/LxwB8tXlsS9ZeURh2WS28qLw/HHSgebLTrCrOnZj8nqlBOyGTa4JS6nDa/HTorXTsvmbu6/tZfV0Q4otKMYM1J1OsfUbJihGOdt2myk9+xSD8lEYyRz2Lv99/M8nnxxFXZ7dN1v0ybRdb/tclPi8vyapvHkA3154t+r+HJOIZGISf8jmnLDlV1JTTnwt2HJ8l28+NpaVq0rIzfHw6VjOzCwX3KsoLDZNCZdq7jonHYsWb6LrGYujuzdJCm6DOaMGs7W/31NpHKnvEiY7JEnsvWTrzDK/ziJwOZx0fmWK+s5pUgEhbO+Z+V9z1C+bhOZ/XrT7a6/kdFbxfU1tLpofq6U6gCsnT17Nm3aJP664WUrdnH1rQsrnMuoadCyuZu3pwyI++aUPf/mNV1XvGhpMdffsahCPo/bxqRrFcMGt4xrNlGREQ7z48njKP5lSbRoaxp2r4cut15Fp+vHsWzSI2yc8iYRf5DUru057Kk7aT70GKtji3q2+c2PWTzhtj/uaWga9hQPx3z9HzIP71Hj59m0aRPDhg0D6Kjr+rrKj8sIG3jvky0EKzUbMk0oKQ2zZPku+vSMbx/mg90A8sxLa2Ie8vv01NWcMKhFQm8oWbG6hOnvbGTDJh89VToXnt2O3Byv1bFqzOZwMOCzaeS9/Sl5b3+KIzOddpedS7NBRwHQ69Fb6fnwRIxgCLvHbXFaYQXTMFh6wwMVb0CbJpGycvTbH6X/x1Pi9lpSsIGdxaHKB1gD0VH2rtIY85T1bPXa0pjXt+8MUe43SPEm5kqL+b9uZ+L9vxMMGpgmrN1QxqxvC3nhkSMSfinfvmxOJ7ljR1bbxlSz2aRYN2LBop2Ei0tiPrbzp0VxfS256QgMGpAVc+lZKGTSu7v1p5xkNYt9ZJLbZat2yZzVTNPk4WdXEggYe/8YRozo2utnpq2xNpxICv78QtY9N4M1j79M2cp1VsepliMjLeaKIQB3q/hOWSbmb3s9O2VoNrmtvBUO0vW4bYwb257Malp91qdLzm9fpTB73DbOPTM3YZs/lfkibC2sepK8aUbn5IXYny1v/Y+vug1j2S0Pod/2KN8eOZIV9zxldayY7G4X7cafU+FQZYgeLtH11qvi+loyJQK43XZeePgIPpmdz1dzCslIdzLmtNb069vU6mgAjDghh5LSMFNnrCMcNtE0OPv0XC69oIPV0arldtmw2TWIVJ1ryki3/sfOCIUoXboaR0YqKR3bWh1H7CNYtIOF4ydh+Cv+wV/z6BSyTxtKZr/DLEpWvR4P3YIRCLLptQ/Q7DY0m40u/7iG1uedFtfXsf43J0F4PHbGnJbLmNMSc4v4uSPbMObU1uwsDpGR4cTlTOw3R06njVOGZvP5VwUVTo/xuG2cP8raApn33ucsuuJ2zHAEMxwmrWcXjnr3WbxtGtbZmMlq6yffoDmq3peJ+INsfvPjhCzYNqeT3s/eQ49/3kywcAfu3Ow6Of1dCnYthMMG/3l/Ix98mkcgYPCn/s2YcFFHmjer2xtPDoeN5lnJc3Pr2gld2FUaYu5PRTidNkIhg5GntOKs0607SHfXYp3fxt2Msc8d/ZKFy/nxlHEcv/iThF5x01iYhgHVrTaOsYkpkTjS03Ck190hG1Kwa+HOh5Yx75fte0eOn39VwI+/7GDGc0eTlir/pHu4XTbum9iLbdsDFBQGaNvaS0a6tfcE1j87HbPSKeNmJIJ/cz7F8xfTpH8fi5KJPVqeOoQlV99V5brd46bVOafWf6AEktjvqxPQ+o2+CsUaIBKJ9nz+ZFa+hckSV/NmbnqpDMuLNUD5xryYW801m41AQaEFiURl7hbNOOzJO7B53NEDh+02bF4P7SacT9OBh1sdz1IyHDxIK9aUxtxS7Q8YLF5WzLlnJv7OzsasxcnHUfTd/ApTIgBGIEiT/n0tSiUqazvubLKGDCDv3c+IlAfIPmPYQe0YbKikYB+kVtnumI2hnE6Ndm3i03ekMTBNk3Ubo/05OrRNqbe547bjzmLd06/i31KA4Y9OjdhTvbS/cizu7Ob1kuFQmIZB+YYtOJtk4GySYXWcOpXSsS2db7rc6hgJRQr2QeqlMmid42XdRh+RfZasOew2zjzFuptpyWT5yhJum/w7xbtCoEFmupP7JvakR7e6L0COtFQG/fgea554mfz3Z+JskkHHay4mZ8zJdf7ahyrvvc9Zcs3dhEvLIGLQ4pTj6DvtnzgzGvdJ8o2JNH+qhZ3FIe5/fDnzf9sBQJtWXiZdq+ilGvaIJx7KfGHGjJtHmS9S4Xpqip13pw2Um7bV2PnTIuYNv6hCvwrN7SLruKMZ8Mk0C5OJeJLmT3WgSaaTh+/sja88QjhsJMTNtGTx1ZxCIjFO3olETL6cU8jIk1tZkCrxrX50CpHyihtJzECQ7d/9jG/9ZlLa127/gG/9Zja//l/Cu0ppOeJ4mg0+WpY2JjAp2Icg2nQpMRsvJaqiHUGCgaqrNAJBg+07q57sIqJ8azcSq0OZze3Cv7mgVgV7y9ufsnD8RMxIBDMYYv1zM2g5YghHzPgXWjW9MYS15LuSZMJhg2/mFvLU1NW8/dGm6DxwEunTM7NCz5Y9PG5b3NvYNiRZx/WPLnGrxAgESe/V9aCfL1xaxqLLJmGU+zGD0Z+hSFk5Wz/9hoKPvzzkvKJu1GqErZTKBKYDGYALuEHX9R/iGUxU5SuPcNXNv7Ilv5xyv4HbZePF19bx5AN96d4l3ep4NXL4YZn06p7BkmW78O8eabvdNnqpDI44rOEWbN/6zRR8OAtMk+xRww96RNzphkvZ9Or7hCIle3f72VO8dLzuEpyZB/+9L/r6x9jbv8t8bPnPx+SMPPGgn1PUvdpOidwAzNZ1/XGllAL+AxwZv1iN07btAd79eDNL9RI6tU/lnJG5tN6n2f+MdzewcXM5wVD0F3bP5p27H17la+NlAAAPY0lEQVTG688nx9yjpmk8cmdv/vt5Hv/7Ih/ThFNPzOHMU1olRf7aWPvMayyf+PDeKY3lt/+L7pNvouPVF9f4OTytsxk0/31W3vMUhbO+x9W8GZ1vHE/rC86oVaZYxXrvY06ZKU1Utf3OPAbsuQPiAPz7+VxRAxs2+5hw468EghFCIZOFS4v5+Is8nri/Lz13L3f74pute4v1vgq2Rbd+57RM7BPI93A4bAndaCuefGs3snziw1U6zy2f9AjZpw4lpVPNG2GltM+l79QH45Ira8jAmNftqV7a/mVMXF5DxN8B57CVUuOVUkv2/Q/oqut6uVIqh+jUyKQ6T9rAPTVlNWW+MKFQdBQWDpuU+w0efmbl3s+xV9f72jQrnMYuEkf+B19AjFUxpmGQ98FMCxJF2T1u+r3zNPZUL/a0FGweNzaPm3aXnUfWCXImZaI64Ahb1/WpwNTK15VSvYE3gJt0Xf+mDrI1KgsW74x5TNnqtaUEQwYup43TT2rFtP+sq3JYcPu2qXXeKVDUjmmamLFaz5mAYW3nueZDj2HY+u8o+HAWoV2ltDhpEGndOlqaSexfrVaJKKV6Am8DY3Vd/zS+kRqnFE/sOUW7Q9s7sj53ZC59emTicdtwOjVSvHaaZjq55xbpsZCockaeGHOJnGa3kTPqJAsSVeTMTKfNxaPpePVFUqyTQG3nsCcDHuCJ6D1HinVdPzNuqRqhUae25vV3N1boAuhyagw/Pntvsymn08a/7unN0hUl/K7vomWWm2P7Z+FM8MMMGrPULu3pdtffWHH3k5ih6O5OzWGn6x3XkNqlvcXpRLKpVcGW4hx/fzm3Hes3+pjzYxFOp0Y4bNK7RwbXXdGlwudpmkYvlSHb4JNI5xsvI/uMYdH5bNMkZ9Rw0lQnq2OJJCTrdxKEw2Hjnlt6klfgZ+2GMtq09tIuV7r/NRRp3TrS5eYJMR8rXbGWnfN+w926Jc2HDkSzy+5ZEZsU7ATTKttDq+zkWJ6XjLbPXcCKOx+n5PeVpHbtQLe7/kbzodasijANg4WXTiTv3c+i66I1DWeTDI75cjopHRpe0zRx6GTyUzQaRd/8yI8jxlH09Y8EC7ezY+4C5p95pWVbsTe+9A7578/E8AeIlPqIlJTh31zAL+deY0kekfikYItGY+nfH6x60ky5n6U3PmBJnvXPvU7EV17xomFQumw15RvzLMkkEpsUbNFolCxZGfO6b+0mjFD9N9GKlMfeIKzZbFULuRBIwRaNiDsn9hFgjsx0NEf9385pdfYIbG5XzDypXTvUex6R+KRgi0ajy8Qrsad4K1yzp3jpdMOlljSe6nTjeLwd2mBPja4G0lwu7KleDn/lIelHLWKSVSKi0Wh3+XmEdu5i9YPPY4QjaDaNDn/7C11uucKSPM6MNAb//AF5b33Cti9/wNs+l3bjz8HbTs4GFbFJwRaNhqZpdLl5Ap2uu4TA1u24WjTDHmNKoj7ZPW7aXDyaNhePtjSHSA5SsEWjY3O58LbJsTqGEAdNJsqEECJJSMEWQtSIaZpEfOWYsfoAi3ohBVuIRsKMRGpdbDe+/C6z2h7L58368UX2ANY8/pIUbgtIwRaNytbPvmHe8L/wde9T+f3GB/DnF1odqc7tWqwz9/gL+MTbi8/S+7DoitsJl5bV+Os3v/k/llx7D8GCIsxIhNCOYlbc+QRrn3y57kKLmKRgi0ZjzRMvseC8ayn6eh5ly1ez/rnX+a7fmQQKtlkdrc74txTww5Cx7Ji7AEwTIxBk84wPmT/qyho/x4q7nqiypT/iK2fVA8/JKLueScEWjUK4zMeKO56osOXbDIUI7dzFmsdfsjBZ3Vr3/OsYgWCFa0YgSPH8RZQsWVGj5/Bvyo95PbSzpMpzi7olBVs0CiVLVkZbmFZiBkMUzvzOgkT1o2TR8phFVbPbKV2xtkbPUd02eXdO85hb60XdkYItGgV3dla1DZ68uQ13TXbmUb2xeaoe0GyEw6T16BLjK6rq8eDfsXkr9mi3pXjo/sBNlmzpb8ykYItGIaVDG5oc1QfN5axw3Z7ipdON4y1KVffaT7gAu9cN+xRWm8dN1vEDSO/RuUbP0eKkwRz1ztNkHN4Dm9dDavdOHP7SQ7S5UE4KrG+y01E0Gv3eeZoF5/2NHfN+Q3M60IDuD08k6/gBVkerM+6WWRz7/dv8fv39FH09D5vXQ9txZ6Huuf6gnqfFSYNpcdLgOkopakoKtmg0XM2aMPCLVynflE9w2w7SenS2vJdIfUjt2oH+H79odQwRB1KwRaPjbZNzUL1EQsUl7PptKe7s5qR1r9k0ghB1oVYFWymVCrwONAPKgIt0XW/4OxBEo7PqwedZef+z2FxOjHCY9J5dOfrDF3C3zLI6mmiEanvT8XLgF13XBwNvALfHL5IQiaHgo9msmvw8hj9AeFcphs/ProXL+OWcq62OJhqpWhVsXdcfB+7f/WE7oCBuiYRIEGsef6nK2YpmKEzxgt8p37DFolSiMTvglIhSajxQ+ZbyOF3X5yulvgR6A8PrIpwQVgpu2xHzuuZ0ENxeLCfDiHp3wIKt6/pUYGo1j52glOoO/A+QuzGiQWk54nh8qzZgBCvuFNQ0jfSe8uMu6l+tpkSUUpOUUhft/rAMiMQvkhCJofNNl+Fs3uSPnYKahj3FS68n/oHN1fCXA4rEU9tlfdOAV3ZPl9iBcfGLJERicDVvxnG/fsS6Z6dT+Nm3eNu0ouO1l9D0mCOsjiYaqVoVbF3XC4BT4pxFiITjataEbrdfTbfb625liH9zAWsem8b2738htWsHOt94GRl9u9fZ64nkJRtnhLCQb81G5gwYQ9hXjhkMUbzgdwo+/IIj33qKlicfZ3U8kWCk+ZMQFlp++6OEdpViBnd3EjQMIj4/i6+6Qw4HEFVIwRbCQkVfzQPDqHI9uLWI4NYiCxKJRCYFWwgLOZtmVvuYPS2lHpOIZCAFWwgLdbxuHPYUb4VrNreL7FHDcaRKwRYVScEWwkLtLj+Pdpefh83twpGZvvtwgf70ef5eq6OJBCSrRISwkKZp9HxkEl0mXkHJstV427YipUMbq2OJBCUFW4gE4GrejKzBzayOIRKcTIkIIUSSkIIthBBJQgq2EEIkCSnYQgiRJKRgCyFEkpCCLYQQSUIKthBCJAkp2EIIkSSkYAshRJKQgi2EEElCCrYQQiQJKdhCCJEkpGALIUSSkIIthBBJ4pDaqyqlugM/Atm6rvvjE0kIIUQstR5hK6UygEeBQPziCCGEqE6tCrZSSgP+DdwK+OKaSAghREwHnBJRSo0Hrq90eT3whq7rC5VSdRJMCCFERQcs2LquTwWm7ntNKbUKGL+7mOcAM4Hj6iShEEIIoJY3HXVd77Ln/5VS64CT4pRHCCFENWRZnxBCJIlDPjVd1/UOccghhBDiAGSELYQQSUIKthBCJAkp2EIIkSSkYAshRJKQgi2EEElCCrYQQiQJKdhCCJEkpGALIUSSkIIthBBJQgq2EEIkCSnYQgiRJKRgCyFEkpCCLYQQSUIKthBCJAkp2EIIkSSkYAshRJKQgi2EEEnikE+cEUI0PEY4zNZPvqZkyQpSO7cne9Rw7G6X1bEaPSnYQogKQjuK+X7w+fg3FxDxlWNP8eK4+UGOnfMW3ratrI7XqMmUiBCigmWTHsa3diOR0jIwDCKlZQQLilh0xe1WR2v0pGALISrIe/tTzGCowjUzEqHoqx8wgkGLUgmQgi2EqCnT6gCiVnPYSikN2ASs3H3pB13XJ8UtlRDCMq3OGcGm1z6oMMrW7Dayhg7E5pIbj1aq7U3HzsACXdfPiGcYIYT1ekz+O9u/+xn/lq1EynzYU1NwZKTS54X7rI7W6NW2YPcDcpVSXwHlwPW6ruvxiyWEsIqzaSbHLfw4uqxvsU5K5/bkjD5JlvUlgAMWbKXUeOD6Spf/CkzWdf1tpdQgYDpwdB3kE0JYwOZwkDPyRHJGnmh1FLGPAxZsXdenAlP3vaaUSgHCux+fo5TKVUppuq7LbQkhhKgjtV0lcidwHYBSqi+wQYq1EELUrdrOYT8ITFdKnUZ0pH1J3BIJIYSIqVYFW9f1HcBpcc4ihBBiP+qql4gdID8/v46eXgghGp59aqY91uN1VbBbAVx44YV19PRCCNGgtQJWV75YVwV7PjAYyAMidfQaQgjR0NiJFuv5sR7UTFMWdwghRDKQ5k9CCJEk6vwAA6VUJtGdkBmAC7hB1/Uf6vp1a0spNRo4R9f1sVZn2ZdSygY8C/QFAsBluq6vsjZV9ZRSA4B/6ro+xOos1VFKOYFpQAfADdyn6/p/LQ0Vg1LKDrwIKKJTjON0Xa8yv5kolFItgV+A4bquL7c6T3WUUr8Cxbs/XKvr+jgr89REfYywbwBm67p+PNH12s/Uw2vWilLqCWAyifnOYxTg0XX9GGAi8KjFeaqllLoZmAJ4rM5yAH8GinRdHwyMAJ62OE91zgDQdf1Y4A7gX9bGqd7uP4IvEO0xlLCUUh4AXdeH7P4v4Ys11E9heozoNxCiI3p/Pbxmbc0FrrI6RDUGAZ8B6Lo+DzjK2jj7tRoYY3WIGngb+Mc+H4etCrI/uq5/AEzY/WF7oMDCOAfyCPA8sMXqIAfQF0hRSs1USn2plBpodaCaiOuUSDWNosbpuj5fKZVDdGrkuni+Zm3sJ+ebSqkhFkSqiQz+ePsGEFFKOXRdT7gio+v6u0qpDlbnOBBd10sBlFLpwDtAwp6Bpet6WCn1CjAaONvqPLEopS4BCnVd/1wplej98X1E/7hMAboCnyqlVCL+Pu0rrgU7VqMoAKVUb+AN4CZd17+J52vWRnU5E9wuIH2fj22J/sOVDJRSbYH3gWd1XX/d6jz7o+v6X5RStwA/KqV66rpeZnWmSi4FTKXUicDhwKtKqZG6rifiDroVwKrdPZBWKKWKiC6n22htrP2rj5uOPYm+9TxP1/WFdf16Ddj3ROcy39r99m2xxXmSnlIqG5gJXK3r+myr81RHKXUR0EbX9clER4YGCbi/Qdf14/b8v1Lqa+DKBC3WEP3j0hv4P6VUa6LvYPOsjXRgdV6wid7E8wBPKKUAinVdP7MeXreheR8YrpSaC2hAUtwkSXC3Ak2Bfyil9sxlj9B1PdFumL0HvKSU+hZwAtfpup7I94KSwVTgZaXUHKKnVV6aDO9YZeOMEEIkiURcviaEECIGKdhCCJEkpGALIUSSkIIthBBJQgq2EEIkCSnYQgiRJKRgCyFEkpCCLYQQSeL/AUxPqYp/fn/HAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X[:,0], X[:,1], c=kmeans.labels_, cmap=plt.get_cmap('coolwarm'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cluster means:\n",
      " [[ 0.16825223  0.70500355]\n",
      " [ 3.11756754 -1.683071  ]\n",
      " [ 0.24825019 -1.28915693]]\n",
      "Clustering vector:\n",
      " [1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 0 0 0 0 2 2\n",
      " 2 0 2 2 0 2 0 0 0 0 0 2 0]\n",
      "Total sum of squares to cluster:\n",
      " 78.52399286287479\n"
     ]
    }
   ],
   "source": [
    "kmeans = KMeans(n_clusters=3, n_init=20)\n",
    "kmeans.fit(X)\n",
    "print('Cluster means:\\n', kmeans.cluster_centers_)\n",
    "print('Clustering vector:\\n', kmeans.labels_)\n",
    "print('Total sum of squares to cluster:\\n', kmeans.inertia_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "80.8614571332361\n",
      "78.52399286287479\n"
     ]
    }
   ],
   "source": [
    "# compare different n_init\n",
    "kmeans = KMeans(n_clusters=3, n_init=1)\n",
    "kmeans.fit(X)\n",
    "print(kmeans.inertia_)\n",
    "kmeans = KMeans(n_clusters=3, n_init=20)\n",
    "kmeans.fit(X)\n",
    "print(kmeans.inertia_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 10.5.2 Hierarchical Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "agg_complete = AgglomerativeClustering(affinity='euclidean', linkage='complete').fit(X)\n",
    "agg_average = AgglomerativeClustering(affinity='euclidean', linkage='average').fit(X)\n",
    "agg_ward = AgglomerativeClustering(affinity='euclidean', linkage='ward').fit(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_dendrogram(model, labels=None, **kwargs):\n",
    "    # Children of hierarchical clustering\n",
    "    children = model.children_\n",
    "\n",
    "    # Distances between each pair of children\n",
    "    # Since we don't have this information, we can use a uniform one for plotting\n",
    "    distance = np.arange(children.shape[0])\n",
    "\n",
    "    # The number of observations contained in each cluster level\n",
    "    no_of_observations = np.arange(2, children.shape[0]+2)\n",
    "\n",
    "    # Create linkage matrix and then plot the dendrogram\n",
    "    linkage_matrix = np.column_stack([children, distance, no_of_observations]).astype(float)\n",
    "    \n",
    "    if labels is None:\n",
    "        labels = model.labels_\n",
    "    else:\n",
    "        labels = [f'{lab1}_{lab2}' for lab1, lab2 in zip(model.labels_, labels)]\n",
    "\n",
    "    # Plot the corresponding dendrogram\n",
    "    dendrogram(linkage_matrix, labels=labels, **kwargs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2IAAAD9CAYAAADed8SZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XuUZFdd6PHvZCBBIUZ8QIioqMBPXAKtAUkwIa0EMAHEB8/hqXLD84ICKmCQyL0qIA9BhDiBOMBlghISooQRlOvkxigsXg3JTfJDUbiijCIaCASRSfr+sU+lq2uqux5ddeqcU9/PWr2m+1RN1T7dtX9n77P3b+9d6+vrSJIkSZLqc9SiCyBJkiRJy8aOmCRJkiTVzI6YJEmSJNXMjpgkSZIk1cyOmCRJkiTVzI6YJEmSJNXMjljDRMTuiHhORHw4ItYi4uqIeFlEHFPDez8pIt49xvN+PSIeNuFrnxMRr6u+f09E/MCI5++LiOdN8h6S6hURt4yIz0XEgUWXZVwRcaeI+HL1/VMj4vkjnr8aEVfVUzpJ8xQRfxYRz+77+a4RsR4Rv9V37HYR8bWIOG4H7/PwiDg45PjN8SQiXhIRTxjxOmO1y9Ret1h0AXSENwC3Be6fmV+MiFsDbwPeCDx+oSXb8OPA1dP+58w8c4ZlkbQ4PwOsAfeKiLtl5jWLLtAkMvPcRZdBUq0OAD8GvKb6+aHAnwIPA15YHftx4IrM/OI8C5KZvz7P11c72BFrkIi4E/BY4A6Z+SWAzPxKRDwV+NHqOccBvw+sAOuUoPLCzDwcEf8JvAo4HbgNcA7wCODuwD8DD61e7zDwUuAM4NbV/79ooCzHUQLV3YFbAu8Hfhl4CnAv4Hci4kbgUuBlwGnAbuBjwLN65d/iPD8NPLwq428Cfw/8YPU+T8nMKwae/2rgHpRAeUJ1/scCd6A0Ah+Vmf8ZEWdWZbmxOn46cEpmfjoifgF4OmUU+AvAMzPz2q3KKGksTwPeDnwKeDbw1IjYD3wkM18JEBFPA1Yz81ER8VDgbOBo4AbgeZn5NxFxDnAypX5/HHgu8AfA7YHjgc8Aj8zMf42Ie1NuWB1dve93A8/JzINbvf5Wha/e99sy85lVXNoH3B/4LuAtmfmigeefAuwHHg18AHg1cBIlHu0CnpyZV0TEtwN/CHwfJd4cAq7KzHMi4m6U2PqtlJj52sw8f+zfuKSdOAC8OCKOysybKB2xFwJvj4jvy8xPUWLApQAR8ZDq8aOB2wFvzswXRcQqpR5/hdKWuTfwa5Q23BeAvx1VkIjYR4kLr6jaby8FHkhp27w8M98w8PyHU9o4ZwKfpcTBu1BiyfXAnszMiLgzcD7wLcDnKLHpf2Xmvoi4b/Uat6a0lX4jMx1xWyCnJjbLicD/HezEZOahzHxn9eNrKZX87pQO0T2B3hS+Y4BDmfkjwJspo2i/CPwAcBylIwPl4n9DZp4IPBI4v2o49Hs1pTF1IvBDwLdRGju/D3wY+OXMvBh4PnAYODEz70np8L10gnO+D/DKzPwhSsPlt/oe21VNZ/xu4MzM/DLw3yiB8CTgzsD3AA+OiG8F3go8LjNXgL8EvgMgIk4DngicWr3Py4GLJyijpAHV9OKTgXdQ4s0Tqnp4HvCkvqc+CTgvIu5Cqd9nVvXwLOCiatQfSj3/ocx8HKWj8zeZeTLwvZRO1eMj4hbARcCLMvMelHi4UpVn1OuP4zaZeSpwX+B5EfE9fef7Y5SO2kMy868psesE4OTM/IHqd9Cb5vhaSiy/G+Vm2H2r17gFcCHw/Cq2nla9z0kTlFHSlDLzk8B/APeIiNsCQbmp8h7gJ6un3R+4NCJ2UW4KPTEz70W56fKCiPi26nk/CDymikU/AfwsJR7dl9LmmsQxwL9l5n0pN6pfHRG36j0YEY+h3Fxfzcyk3Ei/LjNPzsy7Ah8Cnlk9/a3ABZn5g8CzKHGa6nz/EHh8Zv4wpU34hoj4rgnLqhmyI9YsNzH6b3IG8LrMXM/MrwHnVsd6eh22TwFXZuY/VXd9/oFyd6TndQCZ+QngSuB+A+/zEOApEbEGfAT4EUrnb9BDKJX5Y9Vzf4rS8RvXZzJzrfr+owNlfA7ljvuLq3MF+FXg8xHxK5S7QSdQ7kbdD7g6Mz9endebgV6H9sGUTttfV2V8OXDbiOh/L0mTeRrw7sz8QmZ+iBJjzgIOAreKiHtVnbVvp4yoP4Byp/f9VT18GyXm3bl6vQ9k5mGAzHwNpb4+B3g9pcFzG6oYlJkHqn//Eujlb416/XFcUr3uPwH/ykY8uiPwbuBdVcykGmk7mxInX8HGKD+UO9Z7q+d9jtL5ArgrZZTs/KqMlwHfQLnZJakeB4BVStvpz6s20ruBB1Yzk9Yz89rMXKeMmJ0YES+mzDjaRRlNAvjHzPxM9f3pwEWZeX0Vx6YZ5b6k+vejlI5Z733uTelcnZuZ/wiQmRcC+yLiv0fEa6rzuU3V2foRyo14quni769e52RKjHxXFX/eQ5lZdY8pyqoZcWpis3wQuFtEHJuZ1/cORsR3UC7qD6d01Nb7/s9RlCl9PV/r+/7r27zX4YHXuHHg8d3AI3o5HxHxzQPv2/+8Z/caRhFxG+BWQ563la/2fb9OCXI9lwFXUILNSZn5deACyuf2jylTB76r+j+HB/4vlEZYr4xvzcxfrcp4FKUD9x8TlFNSpRplejzwtWpKH8A3Ue7IvgJ4E/AESjx6U2auR8Ru4P2Z+ai+1/lOyij6TwNf7jv+Mkpj4nzK6PYt2bqe92LXdq8/rq3i0WFK5+qSiHhHZn4wIh5MmZr0SkoD6lrgcX3P7y9nfxm/WI3a98p4e2CuuSiSNjkA/ALwn8C7qmPvp3ReTmdjWuKtKekWFwOXU+LRT7FRt7/MZv11/jCT+ypAFS/7X+864DHAH0fEu6t0i6dRbny9jjJd+t8pM4R677tV/LkmM+/TeyAiTgA+P0VZNSOOiDVIZv4z5S7u+RHxTQDVv68HvpCZXwXeCzwzInZFWUnxLODPp3i7J1Sv/8PA91M6Pf3eC/xS3/v8CRvD3ofZ6Pz1ynN01cE5D/jtKcozzIcpQeY6ypA8wIOAl2TmH1U/34cSXK4A7hoR96jO62eBXufxvcBjIuIO1f95Kht3iCRNrpcHcUJm3ikz70SZQngbylS8fZRpPo+gTIWBUuceGBHfD1DldH6CMiI06EHA72bmWykjUw+gakRQOn8/Ub1Gb6R+fcLXn9Shajri84C3RsQ3VmX60yqP48OUBtru6vmXUhp6VNM1f7oqYwJfjYjHVY99J2VE78QZlFHSeP6SMoXwNEr7gKp99RFKO+fS6nl3odxgOjsz/5Qy6nQMG/W83wHgERHxzVVbaJaLq/1tZv5v4PeAt1Sv/yBgX2a+iRJXHgrsrm7iXwH8HEA1vfr+lPjzAeAuEXG/6rEVSi7bd8ywrJqQHbHmeTplRcLeNLoPVj8/uXr8WZSE0Surr6QseDGpH42Ij1Lu8DwqMwdHh55FGRa/ktKYuZIypQ9Kp+y3I+KJwP8APk25a3Q15S7Mc6coz1DV1ICfB55eJZm+ELg4Iq6kJPNfBtw5M/+dcsfoLdV5PYjSYbwhM99HSU7984j4BLAH+JnqtSVN7mnAqzLz5pH0zLyOkhv1S5l5iDK95hPVDSYy82rKjaO3R8THKbHjJ6vcz0EvAV5R1dc/Af6KUs8PU/IwzomIj1FizSFKPZ/k9adSTXm+ljIKdi6wWsWij1Kmg39P1Uj6JeD7q8feSVls5IbM/C/KVO4nV+f2Pkq+2xVHvpukecjMGygdkMzNKyNeSul8Hax+/gRlyuK1EXENpbNzNUOmO2fmeyjtqQ9T2m3zGOX+TUq77JcpMw+eUsWRyykxqFeuJwCPrOLg71Omjd+QmZ+nxM/fqR57KyVf7NNzKKvGtGt93bbosomIdeDbM/PfFl2WWalGDs8GzsnMG6qRvkspd+z9kEsdERG/A7wiM/+lGlH6OPC9VUewESLi6cDHsqwIeQylofTi3hRuSZqXiPg14J2ZeW2UFbA/AZxR3axSw5gjpk7IzC9FxH8BH4qIr1Py4x5pJ0zqnM9QFuT4OhtLxjemE1a5Gvi9Ki/uaOAddsIk1eSTwB9FxE2Udv5L7YQ1lyNikiRJklQzc8QkSZIkqWZjTU2skqJ7iYf/QFkk4TWUxRDel5m/MfD8Yyj7HnyOI5dFl9ROuyl7kHyob1+3hTI2SaKBsQmMT5KAEfFpZEest7N3Zq72HVujrLzy95Tdx384Mz/a99/uTUlOltQ9p1JWsVsoY5OkAY2ITWB8knSEofFpnBGxewLfGBHvq55/DnBMZn4KICLeS9mjoD+YfA7gbW97G8cff/zOii2pEQ4dOsRjH/tYqOp3AxibJDUxNoHxSRKj49M4HbEbKPsVvJGyv8IByga7PddTNvLsdyPA8ccfzx3veMcJiyyp4ZoyZcbYJKlfU2ITGJ8kbTY0Po3TEfsk8HfVMuCfjIgvAt/S9/ixbA4uklQHY5OkpjI+SRppnFUTfx54JUBEnAB8I/CViPi+iNgFPAjnNEuqn7FJUlMZnySNNM6I2JuAfRHxV8A6JbjcBLyNshLI+zLzg/MroiQNZWyS1FTGJ0kjjeyIZeZ/AXuGPHTS7IsjSeMxNklqKuOTpHG4obMkSZIk1cyOmCRJkiTVzI6YJEmSJNVsnMU6OmHvXti/f9GlUFfs2QNnnbXoUmgnjAnNYF2SloMxV2DMH7Q0I2L798Pa2qJLoS5YW/Ni0gXGhMWzLknLw5grY/6RlmZEDGBlBQ4eXHQp1Harq4sugWbFmLBY1iVpuRhzl5sx/0hLMyImSZIkSU2xVCNi6rA6J5+v/W75d/UX63k/J1Srsvcje9l/ZXfmdawdKnVpdV9NdakGe+6+h7NOtL5K6oBZt63m0X5qeRvJjpi6oTf5fGVl7m91cKXGRmNvQn2Lg4xmZ/+V+1k7tMbK8fP/nNdh5fnd6YABrB0q9dWOmKROmHHbaubtpw60keyIqTu6OPncCdUasHL8CgefdHDRxdAQq/tWF10ESZqtJretOtBGMkdMkiRJkmpW64jYIvMbFpmLYM6AJM1XE/LnelMTFz0y5jVHg5pRP5qTE2odUVPU2hFbZH7DonIRzBmQpPlrQv5cE3L3vOZomEbUj4bkhFpH1CS154gtW37Dou+MStKyWLbryzBec7QV60dhHVGTmCMmSZIkSTWzIyZJkiRJNev08vXNSE41eVtaVrOOQfOIJ8YGSW01TYydJo4aJzUvnR4R6yWnLtLK8SsLT+BeO7S28A6ptIxmHYNmHU+MDZLabJoYO2kcNU5qnjo9IgYmp8LiR+OkZdbkGGRskNR2846xxknNU6dHxCRJkiSpiTo/ItZGbcgrAedMS00zaewwV0KS1Bh798L+Cdq/a9W01NXV8f/Pnj1wVnOuYY6INVDT80rAOdNSE00aO8yVkCQ1xv79G52rcayslK9xra1N1tGrgSNiDdXkvBJwzrTUVPOMHdZ7SdJcrazAwYPzee1JRs5q4oiYJEmSJNXMETFpUcaZCz3J/OeGzXuWVP9+lnXvXWnOYD12+jma1efCv7c0W46ISYsyzlzocec/N3Des6T697Osc+9Kcwbrs9PP0Sw+F/69pdlzRExapFnNhW7gvGdJRdNzfqdlzmC9Fv058u8tzZ4jYpIkSZJUMztikiRJklQzpyaqmea9qZ8LW0ittNWiBdstRuACA5L6TbL4iRvfa54cEVMzzXNTPxe2kFprq0ULtlqMwAUGJA2aZPETN77XPDkipuaa16Z+Lmwhtdokixa4wICkYea1+IkxR5NwREySJEmSataaEbFpNjOcdgND5/ZKkpbFtJsF72STYK+zapN55pRZF5Zba0bEptnMcJoNDJ3bK0laJtNuFjztJsFeZ9U288opsy6oNSNiUM9mhs7tlSQtmzo3C/Y6qzaaRx2xLmisjlhE3A74CPAA4DCwD1gHrgKekZk3zauAkrQd45OkJjI2SRplZEcsIm4J/AHw1erQq4CzM/NgRJwLPAy4eH5FlMYwyb5jk+45Bu471lDGp52bV+6DeQ9aZsYmqSbzbP/V0PYbJ0fsFcC5wD9XP58IXFZ9fwA4fQ7lkiYzyb5jk+w5Bu471mzGpx2aR+6DeQ+SsUmqxbzafzW1/bYdEYuIJwGfz8z3RsQLqsO7MnO9+v564Lg5lk8an/uOLRXj0+zMOvfBvActM2OTVLN5tP9qavuNmpr488B6RJwOrABvAW7X9/ixwHVzKpskbcf4JKmJjE2SxrLt1MTMvF9mnpaZq8Aa8ATgQESsVk85A7h8riWUpCGMT5KayNgkaVzTLF//XOC8iDgauAa4cLZFknSEUcmooxJQl2exEeOT5maajY/d9FgVY9OUtqt329Uv64/aYOyOWHVnp+e02RdF0pZ6yahbJZlul3za66R1uCNmfFIdegubTLKJ8TQbHsNGA9OGZLsZm3Zuu3q3Vf2y/qgtWrWhs7TUpk1GdbERaWbq2vjYBU+kDZPWO+uP2mKc5eslSZIkSTPkiJhuNq+NXXucry1p3gbj2LBYZSySNox77R/3um/9ksbniJhuNo+NXXvc4FVSHQbj2GCsMhZJm4177R/num/9kibjiJg2mVf+g/O1JdVluzhmLJKONKtrv/VLmowjYpIkSZJUM0fEauQ8bEmzNCofyhghSbrZqD1JB43ao3SY5dm3dCYcEauR87AlzdJ2+VDGCEnSJr09Sce1srL9PqWD1tYm6+jJEbG6OQ9b0ixtFVOMEZKkI0y7J+k43Ld0Yo6ISZIkSVLN5joidsFVF3DpX1x688/u5yJJ0uic4XFyhb1+StKExs2TGzc/boc5cXMdEbvk2kvcz0WSpAGjcoZH5Qp7/ZSkKYybJzdOftwMcuLmniPmfi6SJB1pJznDXj8laUqzypObQU6cOWKSJEmSVDM7YpIkSZJUM5evl0Ylbo5K2HTzQi2RcTamH3dTenDBCUndtl3MdFEeOSImjUrc3C5h080LtWTG2Zh+nE3pwQUnJHXfdjHTRXnkiJgE0yduunmhlpAb00vS+KaNmcbI7nNETJIkSZJq5oiY2mdYTtdWeVzmb0md0MuzGMypMH9CklpqsD03rC3X8XacHTG1Ty+nqz9va1gOV69Cd7gCS8ui1wnrz6fodcrsiElSCw225wbbckvQjrMjpnYaJ6fL/C2pUwbzLMyfkKSW2649twTtOHPEJEmSJKlmjohp5obtmbHVXhnmd0jLYzA2DIsLxgQtuwuuuoBL/+LSTcesK1I3OSKmmRu2Z8awvTLcH0NaLoOxYTAuGBMkuOTaS0ZeQ60rUjc4Iqa5GGfPDPM7pOWzXWwwJkjFqGuodUXqBkfEJEmSJKlmdsQkSZIkqWZOTWyBSRa/ABN4pXkZVhe3s1093Yr1V5LUeIObMcPwDZmh85sy74QjYi0w7uIXYAKvNE/D6uJ2tqqnW7H+SpJaobcZc7+VleGbMg922HQzR8RaYpzFL8AEXmnexq2L07D+SpJaY7vNmHuWYFPmnXBETJIkSZJq5oiY1DWD87adsy2pRqNyKUflTpon2Q7mr0s754iY1DWD87adsy2pRqNyKbfLnTRPsj3MX5d2zhExqYtGzdt2zrakOZo2l9I8yXYxf13amZEdsYjYDZwHBHAj8HPALmAfsA5cBTwjM2+aXzElaTNjk6QmMjZJGtc4I2IPBcjMH42IVeBVlIBydmYejIhzgYcBF8+tlDPk3HWpMzoVm2alF+MGY5mxS6qNsanFJsl9M65qp0bmiGXmu4Dep+y7gX8BTgQuq44dAE6fS+nmwLnrUjd0LTbNSi/G9ccyY5dUH2NTu42b+2Zc1SyMlSOWmYcj4s3ATwMPBx6SmevVw9cDx82pfHPh3HWpG7oWm2ZlMMYZu6R6GZvabZx2onFVszD2qomZ+UTgrpR5z9/Q99CxwHUzLpckjcXYJKmJjE2SRhnZEYuIx0fEC6ofbwBuAj5czXsGOAO4fD7Fk6ThjE2SmsjYJGlc40xNvAj4w4j4P8AtgV8ErgHOi4ijq+8vnHXBBpMlTZSUNGAhsUlahFHXRK+FjWJskrpq796NfVh7e7b2bwm0Zw+cNX4sHtkRy8yvAI8c8tBpY7/LFPoTzoEtNwgEvPhIS2hRsUlahO2uiV4Lm8XYJHXY/v2lA7ayUr769Tpms+yILdKoZEkTJSVJy2Kra6LXQkmq0coKHDx45PH+kbExjb1YhyRJkiRpNho9IiZ1zlZziyecU6z2c9NQSZotN7RX29gRk+o0bG7xFHOK1X6DOT9gLqwk7cSwuGoMVZPZEZPqNji3eIo5xeoGNw2VpNlyQ3u1iTlikiRJklSz2kfE+vMi+ufwOn9XkqT5GZaXCOYmSmqQJculr31ErD+JcuX4FVaOX2Ht0NrQi4MkSZqN/utvv961uJ/XZUkL0culh418+rW1jc5ZxywkR8z5u5Ik1W+cvETwuixpgZYol94cMUmSJEmqmR0xSZIkSaqZy9dL4+pPIO3pTyTt19GkUkmTc/NuSaMYJ5aTI2LSuPoTSHv6N2bu6XBSqaTJDVskwwUyJPUzTiwnR8SkSQwmkA7T4aRSSdNx825Joxgnlo8jYpIkSZJUM0fEtuBcXWl5ufGtJKm1huW0g3ntDeSI2BacqystLze+lSS11rCcdjCvvYEcEduGc3Wl5eXGt5Kk1honpx3Ma18wR8QkSZIkqWaOiEnLYHC++LB54s4Rl47Qny84LEfQ/EDVYTBv1XxVqRscEZOWweB88cF54s4Rl4bqzxcczBE0P1B1GcxbNV9V6gZHxKRlsd18ceeIS1vaKl/Q/EDVaVTeqp9HqX0cEZMkSZKkmtkRkyRJkqSaOTVxjsZJrjWxVpK0rFyEQm3T+8z2f1b9fGpadsTmqFdRewm1wxJrASuvJGkpjbpOgtdKNcvgZ9bPp3bCjticbZdca2KtJGnZuQiF2qb/M+vnUzthjpgkSZIk1cwRMXVPb/PiwU2L3bBYmpi5rpKk1uq1CXsG24aw0PahHTF1T68TNrhhMdgRkyZkrqskqbUG24QrA3moC24f2hFTNw1uXuyGxdLUzHWVJLXWYJuw34Lbh+aISZIkSVLNWjUiNipXwTwFSZLqZR5h+/T/zfx7SYvTqhGx/g30oEyX6d/Hof9CIEmS5m+7azN4fW6i/r+Zfy9pcVo1IgZb5yqYpyBJ0mKYR9g+tqekxdu2IxYRtwTOB+4EHAP8T+BqYB+wDlwFPCMzb5prKSVpgPFJUhMZmySNa9TUxMcBX8jMU4EzgNcBrwLOro7tAh423yJK0lDGJ0lNZGySNJZRUxPfAVzY9/Nh4ETgsurnA8ADgYtnXzTNmwnWarmFxCfrjaQRlr7tZJyUxrPtiFhmfjkzr4+IYylB5WxgV2auV0+5HjhuzmXUnJhgrTZbVHyy3kjajm0n46Q0rpGLdUTEd1Lu2rw+M/dHxMv7Hj4WuG5ehdP8mWCtNltUfLLeSNqObSfjpDSObUfEIuL2wPuAX83M86vDH4uI1er7M4DL51c8SRrO+CSpiYxNksY1akTshcBtgRdFxIuqY88GXhsRRwPXsHketGrmJtdaYsYnaYbc5HdmjE1qtr17YX/VdlyrppCurm48vmcPnGVdr8O2HbHMfDYleAw6bT7F0aR687B7c68H52ADXjjVScYnabb6ryf91xLwejIJY5Mab//+0gFbWSlf/XodMztitWjdhs46kpsySpJmweuJtCRWVuDgwSOP94+Mae5G7SMmSZIkSZoxR8SkWdhqvrXzrCVtY6u8LPOxJPUYJ7rLETFpFnrzrWFjzvXa2kbnTJKG6N9vqZeb5R5LkvoZJ7rLETFpVgbnWzvPWtIYBvOyzMeSNMg40U2OiEmSJElSzRwRk5aV+4hoSbg/ltQeo/ZHBeususMRMWlZDctr6zG/TR0yLL+ixzwLqVn66ytYZ9VtjohJy8x9RLQk3B9Lao+t6itYZ9UtjohJkiRJUs3siEmSJElSzZyaKDVR/0IacORiGi6kIamjRi3W4EINhb8nqf0cEZOaqH8hDdi8mIYLaUjqsO0Wa3Chhg3+nqT2c0RMaioX0pC0pFxcZTz+nqR2c0RMkiRJkmrmiJjao5c3Zb6UpAXZanNo83EkqSX68/AH25RQa7vSjpjao9cJG9x4GOyISapFLy9nMB8HsCMmSW3Q357sb1NC7e1KO2Jql8G8KfOlJNVsMC/HfBxJapmG5OGbIyZJkiRJNXNEbArmCEjd1Kvb7scjScvF+K9FsCM2BXMEpG7qr9s91m1J6j7jvxbBjtiUzBGQusm6LUnLyfivupkjJkmSJEk1syMmSZIkSTVzaqKWlxtES2oxFxeQtIntmtaxI6bl5QbRklrMxQUkbWK7pnXsiGm5uUG0pBZzcQFJm9iuaRVzxCRJkiSpZo6IaZOtNqvuMfdAktrFXDKpXlu1paxzGmRHTJsM26y6x9wDSWofc8mkeg1rS1nnNIwdMR1hMOegx9wDSWonc8mkelnnNA5zxCRJkiSpZo6INZTzi6XlZf2X1ETGJmm2HBFrqP7E6t4c47VDazcHQEndZf2X1ETGJmm2HBFrMOcXS8vL+i+piYxN0uyM1RGLiPsAL8vM1Yi4M7APWAeuAp6RmTfNr4iStDXjk6QmMjZJGmXk1MSI+BXgjcCtqkOvAs7OzFOBXcDD5lc8Sdqa8UlSExmbJI1jnByxTwE/0/fzicBl1fcHgNNnXah52PuRvazuW2Xt0Bprh9ZY3bfK6r5V9n5k76KLJml6jYhPvfgyGGOML9LSakRskhZi715YXS1fa2vla3W1HNcmIztimflO4Ot9h3Zl5nr1/fXAcfMo2KxttbmeCaZSezUlPpnALqlfU2KTtBD795fOF8DKSvlaWyvHtck0i3X0z2k+FrhuRmWZOxNMpc5bWHwyvkjaRmvbTtJUVlbg4MGNn1dXF1WSRptm+fqPRcRq9f0ZwOWzK44k7YjxSVITGZskHWGaEbHnAudFxNHANcCFsy3SZHqbC/ZvLAi4uaC0nBoVn7qof0NX4IjYC8ZfaQhjk5pt797NUwoa4ogaAAAGEklEQVR7I1h79sBZxvN5GasjlpmfBk6qvv8kcNocyzSR/tyvnl7DwIaA1H1Njk9dNBhz+2MvGH+lHmOTWqXXCVvpi+m9TpkdsbnpxIbO5mZIUn0GY24/468ktZR5XbWbJkdMkiRJkrQDnRgRk+Zq2Lxp50xL2kZ/Ll1/Hp35c5J6jBOyIyaNMjhv2jnTkkbYau9KMH9OUmGckB0xaRz986adMy1pDOYvSxrFOLHczBGTJEmSpJrZEZMkSZKkmjk1UbUyMVVSP2OCJKm1egu6wVSLujkiplr1ElOBm5NT1w6t3dwQk7RcjAmSpNbqX1V7ZaV8ra1tdM5GcERMtTMxVVI/Y4IkqbV2sBG2I2KSJEmSVDNHxGpmPoQkqS161yyvV9op2z/SkeyI1czN+yRJbdF/zQKvV5qe7R/pSHbEFsB8CElSW/Rfs7xeaSds/0ibmSMmSZIkSTVzREzSht5+GP17YUDZD+PMMxdWLEnLbViuGsCeu+/hzNsbm3q2+z05/U9qHjtikjb0OmErKxvHep0yO2KSFmQwVw028ovsiG3Y7vdkR0xqHjtikjbbwX4YkjQv5heNx9+T1B7miEmSJElSzeyISZIkSVLNnJootcV2C2mc5dx/aRIuaiC1h5tBq6vsiEltsd1CGnbEpIm4qIHUHm4Gra6yIya1iQtpSDPjogZSe1hf1UXmiEmSJElSzRwRG2DegKR5Mc9BktRqw/LVzVWfmh2xAeYNSJoX8xwkSa02mK9urvqO2BEbwnnIkubF+CJJarX+fHVz1XfEHDFJkiRJqpkjYpoJc1+kdjEfVpLUWh3ZW9WOmGbC3BepXcyHlSS1Vkf2VrUjppkx90VqF+usJKm1OrC3qjlikiRJklQzR8Q6yNwPSdIimTfcPrYdpPrZEesgcz8kSYtk3nD72HaQ6mdHrKPM/ZAkLZLXofbxbybVa6qOWEQcBbweuCfwNeDJmfl3syyYJE3D+CSpiYxNkgZNu1jHTwG3ysyTgecDr5xdkSRpR4xPkprI2CRpk2mnJp4C/BlAZn4gIu418PhugBu/dCMAn/3sZ29+4PB1h1t/rCnlqOsYwKMvfDRXf/5qAE753VN4+8Pfvunxmb/n4XKMJh5rSjlqPnbo0KHekd0023bxaTeUcxl6qtb/qc4fSowAao0TTTn/ZTrWlHL0H+tIbIIWtJ2aUo4mHWtKOcpB20RTHwN49KPh6nIN45RTyr9v33wdm/T8R8WnXevr68OObysi3gi8MzMPVD//P+B7M/Nw9fMpwOUTv7CkNjg1M/9q0YXYynbxydgkdVprY1P1s/FJ6q6h8ekWU77Yl4Bj+34+qhdIKh8CTgU+B9w45XtIapbdwB0o9bvJtotPxiape7oQm8D4JHXRtvFp2o7YFcBDgT+OiJOAK/sfzMyvAY29KyVpap9adAHGsGV8MjZJndXq2ATGJ6nDtoxP03bELgYeEBF/DewCfm7K15GkWTM+SWoiY5OkTabKEZMkSZIkTW/aETFpLiJiN3AccF1m3rRsx5pSjq2OdVVTfsdNKYfn7/m37fylZdaUOjzrY00px06PbccRMTVGRDwdeDDwH8BtgYuAY5bo2OC/TSrbRZn5pm3+fK3VoM9do//+nr/n39Tz72psksbRoGtYY2JCk46Nik9zGRGLiI8CX66+dgHrwPENPnZf4OMNKMeyn//dMvNOVCLiDcCNmfngJTl2ZWbevQHl2OpYJxo7Q+JTUz53Tf/7e/6ef1PPv6uxaZmu/0061rbfSVOuYU2KCU06Vn9HDHgk8AuZ+YK+wty5qceaUo5lP/+IuCDKSlL/CNwR+CbgqCU6dmxDyrHVsa7YFJ8a9Llr+t/f8/f8m3r+XVF726kp1/8mHWtKOSYob1OuYU2KCU06tq25TU2MiNtn5r+05VhTyrHM5x8RJwBPodzt+QzlLsLuJTr2buBnG1COoccG/3Zt1tDPXaP//p6/59/U8+9qbKrrWBOu/0071pRyjHOsQdewxsSEJh0bFZ/MEZMkSZKkmh216AJIkiRJ0rKxIyZJkiRJNbMjJkmSJEk1syMmSZIkSTX7/wdh5GSOIP5RAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x288 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2, ax3) = plt.subplots(1, 3, figsize=(15,4))\n",
    "\n",
    "color_threshold = 48\n",
    "\n",
    "ax1.set_title('Complete linkage')\n",
    "ax2.set_title('Average linkage')\n",
    "ax3.set_title('Ward linkage')\n",
    "\n",
    "plot_dendrogram(agg_complete, color_threshold=color_threshold, ax=ax1)\n",
    "plot_dendrogram(agg_average, color_threshold=color_threshold, ax=ax2)\n",
    "plot_dendrogram(agg_ward, color_threshold=color_threshold, ax=ax3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# with scaling\n",
    "agg_complete_scaled = AgglomerativeClustering(affinity='euclidean', linkage='complete').fit(StandardScaler().fit_transform(X))\n",
    "plt.title('Complete linkage, scaled features')\n",
    "plot_dendrogram(agg_complete, color_threshold=color_threshold)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# correlated distances\n",
    "X3 = np.random.normal(size=(50,3))\n",
    "\n",
    "corr_X = pairwise_distances(X3, metric=\"correlation\")\n",
    "agg_corr = AgglomerativeClustering(affinity='precomputed', linkage='complete').fit(StandardScaler().fit_transform(corr_X))\n",
    "plt.title('Correlated distances')\n",
    "plot_dendrogram(agg_corr, color_threshold=color_threshold)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 10.6 Lab 3: NCI60 Data Example"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pandas.io.json import json_normalize"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>data</th>\n",
       "      <th>labs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{'1': 0.30000000000000004, '2': 1.18, '3': 0.5...</td>\n",
       "      <td>CNS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>{'1': 0.67996, '2': 1.29, '3': 0.1699600000000...</td>\n",
       "      <td>CNS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>{'1': 0.9400000000000001, '2': -0.04, '3': -0....</td>\n",
       "      <td>CNS</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>{'1': 0.28, '2': -0.31, '3': 0.68, '4': -0.81,...</td>\n",
       "      <td>RENAL</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>{'1': 0.485, '2': -0.465, '3': 0.395, '4': 0.9...</td>\n",
       "      <td>BREAST</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                data    labs\n",
       "0  {'1': 0.30000000000000004, '2': 1.18, '3': 0.5...     CNS\n",
       "1  {'1': 0.67996, '2': 1.29, '3': 0.1699600000000...     CNS\n",
       "2  {'1': 0.9400000000000001, '2': -0.04, '3': -0....     CNS\n",
       "3  {'1': 0.28, '2': -0.31, '3': 0.68, '4': -0.81,...   RENAL\n",
       "4  {'1': 0.485, '2': -0.465, '3': 0.395, '4': 0.9...  BREAST"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ncl60_raw = pd.read_json('Data/NCI60.json', orient='records')\n",
    "df_ncl60_raw.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    {'1': 0.30000000000000004, '2': 1.18, '3': 0.5...\n",
       "1    {'1': 0.67996, '2': 1.29, '3': 0.1699600000000...\n",
       "2    {'1': 0.9400000000000001, '2': -0.04, '3': -0....\n",
       "3    {'1': 0.28, '2': -0.31, '3': 0.68, '4': -0.81,...\n",
       "4    {'1': 0.485, '2': -0.465, '3': 0.395, '4': 0.9...\n",
       "Name: data, dtype: object"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ncl60_raw.data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 64 entries, 0 to 63\n",
      "Data columns (total 2 columns):\n",
      "data    64 non-null object\n",
      "labs    64 non-null object\n",
      "dtypes: object(2)\n",
      "memory usage: 1.1+ KB\n"
     ]
    }
   ],
   "source": [
    "df_ncl60_raw.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>10</th>\n",
       "      <th>...</th>\n",
       "      <th>6821</th>\n",
       "      <th>6822</th>\n",
       "      <th>6823</th>\n",
       "      <th>6824</th>\n",
       "      <th>6825</th>\n",
       "      <th>6826</th>\n",
       "      <th>6827</th>\n",
       "      <th>6828</th>\n",
       "      <th>6829</th>\n",
       "      <th>6830</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>labs</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CNS</th>\n",
       "      <td>0.30000</td>\n",
       "      <td>1.180</td>\n",
       "      <td>0.55000</td>\n",
       "      <td>1.14000</td>\n",
       "      <td>-0.26500</td>\n",
       "      <td>-7.000000e-02</td>\n",
       "      <td>0.35000</td>\n",
       "      <td>-0.31500</td>\n",
       "      <td>-0.450000</td>\n",
       "      <td>-0.654980</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.99002</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.03000</td>\n",
       "      <td>-0.17500</td>\n",
       "      <td>0.62998</td>\n",
       "      <td>-0.03000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.28000</td>\n",
       "      <td>-0.34000</td>\n",
       "      <td>-1.930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CNS</th>\n",
       "      <td>0.67996</td>\n",
       "      <td>1.290</td>\n",
       "      <td>0.16996</td>\n",
       "      <td>0.37996</td>\n",
       "      <td>0.46496</td>\n",
       "      <td>5.799600e-01</td>\n",
       "      <td>0.69996</td>\n",
       "      <td>0.72496</td>\n",
       "      <td>-0.040039</td>\n",
       "      <td>-0.285020</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.27006</td>\n",
       "      <td>-0.30004</td>\n",
       "      <td>-0.25004</td>\n",
       "      <td>-0.53504</td>\n",
       "      <td>0.10994</td>\n",
       "      <td>-0.86004</td>\n",
       "      <td>-1.25</td>\n",
       "      <td>-0.77004</td>\n",
       "      <td>-0.39004</td>\n",
       "      <td>-2.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CNS</th>\n",
       "      <td>0.94000</td>\n",
       "      <td>-0.040</td>\n",
       "      <td>-0.17000</td>\n",
       "      <td>-0.04000</td>\n",
       "      <td>-0.60500</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.09000</td>\n",
       "      <td>0.64500</td>\n",
       "      <td>0.430000</td>\n",
       "      <td>0.475020</td>\n",
       "      <td>...</td>\n",
       "      <td>0.31998</td>\n",
       "      <td>0.12000</td>\n",
       "      <td>-0.74000</td>\n",
       "      <td>-0.59500</td>\n",
       "      <td>-0.27002</td>\n",
       "      <td>-0.15000</td>\n",
       "      <td>0.00</td>\n",
       "      <td>-0.12000</td>\n",
       "      <td>-0.41000</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>RENAL</th>\n",
       "      <td>0.28000</td>\n",
       "      <td>-0.310</td>\n",
       "      <td>0.68000</td>\n",
       "      <td>-0.81000</td>\n",
       "      <td>0.62500</td>\n",
       "      <td>-1.387800e-17</td>\n",
       "      <td>0.17000</td>\n",
       "      <td>0.24500</td>\n",
       "      <td>0.020000</td>\n",
       "      <td>0.095019</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.24000</td>\n",
       "      <td>-0.11000</td>\n",
       "      <td>-0.16000</td>\n",
       "      <td>0.09500</td>\n",
       "      <td>-0.35002</td>\n",
       "      <td>-0.30000</td>\n",
       "      <td>-1.15</td>\n",
       "      <td>1.09000</td>\n",
       "      <td>-0.26000</td>\n",
       "      <td>-1.100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BREAST</th>\n",
       "      <td>0.48500</td>\n",
       "      <td>-0.465</td>\n",
       "      <td>0.39500</td>\n",
       "      <td>0.90500</td>\n",
       "      <td>0.20000</td>\n",
       "      <td>-5.000000e-03</td>\n",
       "      <td>0.08500</td>\n",
       "      <td>0.11000</td>\n",
       "      <td>0.235000</td>\n",
       "      <td>1.490000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.55498</td>\n",
       "      <td>-0.77500</td>\n",
       "      <td>-0.51500</td>\n",
       "      <td>-0.32000</td>\n",
       "      <td>0.63498</td>\n",
       "      <td>0.60500</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.74500</td>\n",
       "      <td>0.42500</td>\n",
       "      <td>0.145</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 6830 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           1      2        3        4        5             6        7     \\\n",
       "labs                                                                       \n",
       "CNS     0.30000  1.180  0.55000  1.14000 -0.26500 -7.000000e-02  0.35000   \n",
       "CNS     0.67996  1.290  0.16996  0.37996  0.46496  5.799600e-01  0.69996   \n",
       "CNS     0.94000 -0.040 -0.17000 -0.04000 -0.60500  0.000000e+00  0.09000   \n",
       "RENAL   0.28000 -0.310  0.68000 -0.81000  0.62500 -1.387800e-17  0.17000   \n",
       "BREAST  0.48500 -0.465  0.39500  0.90500  0.20000 -5.000000e-03  0.08500   \n",
       "\n",
       "           8         9         10    ...       6821     6822     6823  \\\n",
       "labs                                 ...                                \n",
       "CNS    -0.31500 -0.450000 -0.654980  ...   -0.99002  0.00000  0.03000   \n",
       "CNS     0.72496 -0.040039 -0.285020  ...   -0.27006 -0.30004 -0.25004   \n",
       "CNS     0.64500  0.430000  0.475020  ...    0.31998  0.12000 -0.74000   \n",
       "RENAL   0.24500  0.020000  0.095019  ...   -1.24000 -0.11000 -0.16000   \n",
       "BREAST  0.11000  0.235000  1.490000  ...    0.55498 -0.77500 -0.51500   \n",
       "\n",
       "           6824     6825     6826  6827     6828     6829   6830  \n",
       "labs                                                              \n",
       "CNS    -0.17500  0.62998 -0.03000  0.00  0.28000 -0.34000 -1.930  \n",
       "CNS    -0.53504  0.10994 -0.86004 -1.25 -0.77004 -0.39004 -2.000  \n",
       "CNS    -0.59500 -0.27002 -0.15000  0.00 -0.12000 -0.41000  0.000  \n",
       "RENAL   0.09500 -0.35002 -0.30000 -1.15  1.09000 -0.26000 -1.100  \n",
       "BREAST -0.32000  0.63498  0.60500  0.00  0.74500  0.42500  0.145  \n",
       "\n",
       "[5 rows x 6830 columns]"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get data\n",
    "df_ncl60 = json_normalize(df_ncl60_raw.data)\n",
    "# rename columns so they are integers and not strings\n",
    "df_ncl60.rename(columns=lambda name: int(name), inplace=True)\n",
    "# sort columns \n",
    "df_ncl60 = df_ncl60.reindex(columns=sorted(df_ncl60.columns))\n",
    "# each row is a type of cancer\n",
    "df_ncl60.index = pd.CategoricalIndex(df_ncl60_raw.labs)\n",
    "df_ncl60.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RENAL          9\n",
       "NSCLC          9\n",
       "MELANOMA       8\n",
       "COLON          7\n",
       "BREAST         7\n",
       "OVARIAN        6\n",
       "LEUKEMIA       6\n",
       "CNS            5\n",
       "PROSTATE       2\n",
       "UNKNOWN        1\n",
       "MCF7D-repro    1\n",
       "MCF7A-repro    1\n",
       "K562B-repro    1\n",
       "K562A-repro    1\n",
       "Name: labs, dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ncl60.index.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1,  1,  1, 12,  0,  1,  1,  0,  9,  9, 12, 12, 12, 12, 12, 12, 12,\n",
       "        0,  9, 12, 13, 10,  8, 11, 10, 10, 10, 10, 10, 11,  9,  9,  9,  5,\n",
       "        4,  3,  5,  5,  5,  5,  5,  2,  2,  2,  2,  2,  2,  2,  6,  0,  7,\n",
       "        0,  9,  9,  9,  8,  0,  0,  8,  8,  8,  8,  8,  8], dtype=int8)"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ncl60.index.codes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 10.6.1 PCA on the NCI60 Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "scaled_pca = make_pipeline(StandardScaler(), PCA())\n",
    "pca = scaled_pca.named_steps['pca']\n",
    "pcaX = scaled_pca.fit_transform(df_ncl60)\n",
    "pcaX = pcaX[:,:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10,4))\n",
    "\n",
    "ax1.scatter(pcaX[:,0], -pcaX[:,1], c=df_ncl60.index.codes, cmap=plt.get_cmap('rainbow'))\n",
    "ax1.set_xlabel('Z0')\n",
    "ax1.set_ylabel('Z1')\n",
    "ax2.scatter(pcaX[:,0], pcaX[:,2], c=df_ncl60.index.codes, cmap=plt.get_cmap('rainbow'))\n",
    "ax1.set_xlabel('Z0')\n",
    "ax1.set_ylabel('Z2');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlgAAAEBCAYAAACkD9t+AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl81OW1+PHPzGQPSYCQhbAj5LCKCIiKIIpoLbWiXr3urb1Ya9Vri71drPZqW3+tvRaqrVtdat1w360ioGGVRWWRkDwJS0gge0Iy2SbLzPz+mBkakeAkmWRmkvN+vXjlOwN+v8dXy3jmPOc5j8XtdqOUUkoppQLHGuwAlFJKKaX6Gk2wlFJKKaUCTBMspZRSSqkA0wRLKaWUUirANMFSSimllAqwiGAH4CMi0cAsoARwBjkcpVTPswFDgW3GmOZgB9Md+vmlVL90ws+wkEmw8Hw4rQ92EEqpXjcX2BDsILpJP7+U6r+O+xkWSglWCcALL7xAenp6sGNRSvWw0tJSrrnmGvD+3Q9z+vmlVD/zTZ9hoZRgOQHS09MZPnx4sGNRSvWevrCkpp9fSvVfx/0M0yZ3pZRSSqkA0wRLKaWUUirANMFSSimllAowTbCUUkoppQIslJrc/eJ0ucky5WQX25mckch8ScVmtQQ7LKVUmBKR2cD9xpj5IjIOeAZwA7uBW4wxLhH5X2AR0Ab8xBizNWgBK6W65Nj8Ye74FNbnV5BdbGdiegJYIKekLmC5RVglWE6Xm5ue+4xSu4O541NYvjqPFVsLefy6mZpkKaU6TUR+DlwHNHjfWgbcZYzJEpHHgItF5CBwNjAbGAG8jmfulVIqxLRPotonTRPTE3hpWxFldZ78Ydkqwy9e30VaYgxzxiWz9JWdAFw9e2TAcouwWiLMMuWU2h3ceeFEVmaX8sKS2ZTUOsgy5cEOTSkVnvYBl7Z7PQNY673+ADgPOAv4yBjjNsYUAhEiktK7YSqljsfpcrMmp4yH1uSzKruUHz77GctX59HQ0sbSV3ey9JWdNDS3cc+7e9i0r4pLpw+judWFBQtV9S3sKbbz93UHqGtuo665jXd2FvPmj+cEJLcIqwpWdrGdueNTKLU72F/RQE1jK/MyU9hTbGfBxLRgh6eUCjPGmNdFZHS7tyzGGLf3ug5IAhKBqnZ/xvd+Ra8EqZQ66tgKVfuq1O/e30NVfQu/u3gKn+SVYwEcrU6e3niAVqfnr/Vv38shNtJGfHQESbGRZKYlYLNaaHO5AEhNiCHSZg1IbhFWFazJGYmsz68gOsIGQG1jK+vyKpiUkRjkyJRSfYSr3XUCUAPYvdfHvq+U6gW+KtVfVudx6SMbWbYqj6ZWJ/e8u4eN+yqZMXIQnxccoaTGQUOLk6Wv7uS9nZ7h6ulJMUwdlsR1p48kKTaCJWeNYc9vL+C+S6bgaHPyX3NHs2TuGBpbnDQ0t3HpqcNodboCkluEVQVrvqSyYmshf1qZC8DtL+/gpJR45ktqkCNTSvUR20VkvjEmC7gQ+ATYC/xJRB4AhgNWY0xlEGNUqk87bpXK7iB5QDR7SuwMjovi7e2HKa51APDClkJOHp7E3MwhbD1whIunDeWszBT+9vFe3G43t5wzjjnjhvDW9mI+2lNGZISVtaacAdERPLRmL2eNH0JhVSMA2wqqWbYqj6FJMd3OLcIqwbJZLTx+3UyeXL+fP3yQy+JTMrj13PHa4K6UCpQ7gCdEJArIAV4zxjhFZD3wKZ6q/y3BDFCpvqajZb8Zowbzyze+pLapleT4SHYX2wGobGhh+oiBzMtM4cPdpVx3+ijuuEBwtDqZ9fvVrN9bxYDYyK8lTaeNGcSVp40kt6SOO86Xo7sI9xTbWXbFNLBAbkkdSxdm9r9dhOBJsuZlpvCHD3KR9ARNrpRS3WKMKQBO917n4dkxeOyfuQe4pzfjUqov8yVVXx6u5ZPcctpcbuaOH8Jdb+/mSEMLY4fE889NBQBYgOGD41g4OZ0t+6s5R1JIiInkxnljeX9XCW/vLKbN7WZdXsVXkqiOkqaFk/59IPuCiWlf6bNq/3vdFXYJFkBclKcHq7GlL5wRq5RSSvVtHVWpMpJi2VNiJzk+ipe3FXGksdXz54Gff0uIttl4dO1e5o5L4bYF47nx2c94dvNBzpVULnt001cSqo6SqEAmTZ0RpgmWJ+wGTbCUUkqpkHOi3X73vruHyvpmzjgpmTW55ThdbsrrmjltzGCmDR/Ia58fYtGUofx4/jgcrU7+sjqPN7YfptnporS2CUlLQNITuWzG8OMmVKEiLBOs+GhvBau5LciRKKWUUqq9Y4eC/+79PVQ1tHDbueP5OLecQzVNAOwsqmF+Zgr7Kuo5b2IaSbGeZb8XtxQeTaiOXfa743wJmxNcwjLBivGOadAKllJKKRUafFWrd3cWk19Wzz9umMWanHJa2lw0NDv54we5SFoCF508lPV7K7nujNH8t3fZ7/ktnVv2CwdhmWBZrRbiomw0tWgFSymllAqG4y0DHq5porGljeIaB+f+2XMowsjBscRF2bh8xnDuvXjK0d1+b24/TEuYLft1RlgmWODpw9IKllJKKdU7OuqrOm30YH7xxi5qG1vxDkwnOsLKoLhIlp6fySXThzPr96v5xFQQ92FuWC/7dUYYJ1g27cFSSimlesGxfVW/fX8PFXXNnDF2MM9+epA2lxurBS46eSi3nDOOBz4y7Ciq4c0vDvP85sI+s+zXGeGdYGkFSymllOox7fuq8srq+dNlU3lnVwlltQ5anG62F9bwg7PGkJ4YwyNZezkpZQAThibyt6tP5fzla8kYGMtF0zL6RUJ1rLBNsOKjIzTBUkoppQLoeMuAxTVNNDS3cbimiSuf2EJMhJVTRw0ip8TO9WeMYen5mThanSxflfeV3X+ZaQn8+YpT+tzSn7/CNsGKi7JRr0uESimlVEC0XwY8a9wQfv3WbqobWsDtps0NsZFWBsZG8utFkzhvUhqzfr+at3YcptX19XEKgTpuJpyFdYJVUdcc7DCUUkqpsNW+YtXmdHHoSBMXTcvglc+KKPf+N3bBhFR+cl4mD67JY0dRDc9+WsCja/f1y76qzgjbBCs+KoIGHdOglFJKdcrxzgGUtAT+9WUJjjYXuaWG00YPZuGkNN7afpipw5KYOjxJ+6o6KWwTrLhoG43N2oOlVH8nIvM6+j1jzLrejEWpUNd+GdB3DmC0zUp2sZ3oCCs2q4W7Fk3khjljcLQ6eXlrkfZVdVH4JlhawVJKedzs/XkSEAVsA6YD9cD8IMWkVMg4dhmwsLqReeNTeG7zQVqdbqwWN9eePpL/uWAC3/3bBh7N2kd5XbP2VXVTGCdYNhytLpwut/6PrVQ/Zoy5CkBE3gcuNsa0iYgNeD+4kSkVHMcdCGp3cFLqAD7cXUpzm4u95fVMHzGQ6sZWFk5MJSEmkrgoG/FRNi6elkGkzap9Vd3UYwmWiMwG7jfGzBeRccAzgBvYDdxijHF15/7xUZ7Qm1qdDIgO2zxRKRU4Q9tdRwCpwQpEqWA53kHL5fZmRifH8faOYuKiPMuA9y2ewuUzR3Djs5/x7GbPOYCLH95IxsBYbj8vUwsXAWDtiZuKyM+BJ4EY71vLgLuMMXMBC3Bxd58RG+U58FmnuSulvJ4CskXkdWAH8Mcgx6NUr3G63KzJKeOOV3aQX1bP49fOIDrCSkVdM442F/bmNv5w6VS2/XohwwfFsmxVHg98ZL5yDuDShZk8ft1MTa4CpKdKP/uAS4HnvK9nAGu91x8A5wNvducB8dHeBEuHjSqlAGPMwyLyHDAB2G+MqQx2TEr1hvZVKwtQZndw1p8+we2Gk4cncbCqgctnjOCq00bS6nR9ZRmwr54DGAp6JMEyxrwuIqPbvWUxxniPgKQOSOruM+K8S4Ta6K6UAhCRycBjwEDgBRHZbYx5L8hhKdUjjm1c31fRwJghcXycWwHAwNgIfnbBBP5jhueg5Te3H6bFuxNQlwF7R281L7Xvt0oAarp7Q18PllawlFJeDwE3AE/gWS78ANAES/U5RytWtQ7GpsTzwe5SWp1uqhuaue3ccewoqiGnxM47Ow6zYmvHBy2rntVbCdZ2EZlvjMkCLgQ+6e4NfT1YDdqDpZTyMsbsFRG3MaZCROqCHY9SgdK+YtXqdJJfXk9yfBTv7CwhMSYCt9vJ7y6ewndPGYaj1akDQUNAbyVYdwBPiEgUkAO81t0b+nqwmrSCpZTyqBaRm4B4EbmSAFTKlQqm401cH50cx8rsMtpcblpanfzu4slceupwvv3Qen7/fg45pXU6EDRE9FiCZYwpAE73XucBZwfy/vFHe7A0wVJKAfBfwJ1AJTDT+1qpsNS+cX3EoFhySuxER3omrg+Oi8TuaOOu70xi0ckZX2tc12XA0BC2A6SOjmnQJnelFGCMsYvIMv49HmYAUB3EkJTqlGMb10tqHdy1aCL/8+ouWpxuXG4X150+kju/PYlvPbiOe9/dw+5iuzauh6iwTbC0yV0p1Z6IPIKnx7MEz7w9N3BmUINSyk/HDgh9eWsRja1tXPXEFhJjIxiaFMO3p6QzMC6KCJtFK1ZhIGwTrJhIKxaLDhpVSh11GnBSd0+JUKq3HK9i9dcrp/Pgx/lUN7YAcOmpw/jtd6fw3y9t54WthTpxPYyEbYJlsViIi7RpD5ZSymcvnuXBxmAHotQ3ObZi9epnRTS3uTj/L2uJsFm56eyx/GtXKRvyK3k4a+9XJq5fNmO4VqzCQNgmWABx0RHag6WU8hkJHBSRvd7XbmOMLhGqkOKrWr27s5j8snrevnUOL24tpN7RhqPNxZxxyfz58lNIHhDF+rwKLpmoE9fDVVgnWPFRNu3BUkr5XBXsAJQ6kfZVq9hIG9WNLcz+f2tobnOxYEIKOaV15JfV889PC7RxvQ/okcOee0tcVAQNzZpgKdWficgS7+WPgJuO+aVUyMgy5ZTaHfzqwglU1DVT52gD3PzP+cJj181kYGwkV84aQVykTQ9e7gPCuoIVF2XTJUKlVJH3Z25Qo1DqONo3su86VEODo41rntzK0MRoJmckUmZ3sDavnH/tLtGKVR8T3glWdAT2ptZgh6GUCiJjzErv5QvALCASz5iGjKAFpRT/XhI8XNNEdISNHUWewwVuXzCOm+ePAzjukTaqbwjrBCs+ykZpbVOww1BKhYY3gChgGGADioEVQY1I9TvtK1YtTic5JXU4Wtuobmzl0ukZrMwu483txbQ43XqkTR8X1glWbJRNe7CUUj5JxpizReRJ4DZgVbADUv1L+yb2CemJvLX9MG0uN6eMSOKZG6YydXgSf/ggh7Jax9E+K61a9V1hnWDFR0XQ1KoJllIKAF+/QLwxpsl7uLxSPa796IXc0jrOGpfMy58dYkCUDUebi5vOPompw5NodbrYkF/J0oWZLJiYFuywVQ8L6wQrLtpGg05yV0p5vCkivwF2ishmoC7YAam+72jVqtZBQ3Mbh4808dK2Q1x/xih+cl4mlzyykf99O5tdh2pZl1fB0KQY5ktqsMNWvSC8E6zICJrbXLQ5XUTYwnrihFKqm4wxD/uuReR9IL8r9xGRSOCfwGjACdwItAHP4DnfcDdwix7J038de8TNgcoGEmMiOFDVSGykjaFJ0ZydmUJCTISeGdiPhXWCFR9tA6Cx1UmiJlhK9UsisgJP4nM8V3fhlt8GIowxZ4rIQuA+PDsT7zLGZInIY8DFwJtdCliFtfZ9VmeMTeaFLYU0tjgZFBfJ/ZdN5aPsMnYequHxtftYtipPRy/0Yx0mWCIyr6PfM8as65lwOicuyhN+U4uTxJjIIEejlAqSxwJ8vzwgQkSsQCKe3q7TgbXe3/8AOB9NsPol37DQny7M5J53smlscWKzwN2LJnHpjOFcfMowHb2ggBNXsG72/jwJz9bnbcB0oB6Y37Nh+cdXwdI+LKX6L2PMWgARSQV+DWQC2XgqT11Rj2d5MBcYAnwHmGeM8VXJ6oCkboSswpBvWfCxtfuorm/hv575jLEp8by4ZDa/evNL/vhhLvkV9Tp6QR3VYYJljLkKjvYyXGyMaRMRG/B+bwX3TWIjvUuEeh6hUgpe9v56GpgDPIcnOeqsnwIrjTG/EpERwMd4vmT6JAA13YxVhRGny80Pn/uM3JI6yuwO2lxuRifH8d5tZxFps2qflTouf3qwhh7z50Nm+0N8tCd8TbCUUgDGGN9y4U4RuaKLtznCv0c+VOPpv9ouIvONMVnAhcAn3QpUhbz2jezVDS1s2V9FfbOT6SOSiIqwsfNQDT97ZScHqxu1z0odlz8J1lNAtojsBiYB/9uzIfkvLsq7RKjnESqlIFdErsGT/MwAqkQkE8AYk9eJ+ywHnhaR9XgqV3cCnwFPeGdr5QCvBTRyFVLaj14YPCCK9XmVAPzmOxP5/pljcAO3vvgFtU2tWrFSHfrGBMsY87CIPAdMAPYbYyp7Piz/+JrcG3Wau1LK8xk1AVjS7r3H8ewwPNffmxhj6oHjVb/O7lZ0KqQdO3qhsLqR5AFRrMurRNIGUFDVyLBBsVitFlq9v68DQ9WJfGOCJSKT8ezSGQi8ICK7jTHv9XhkftAKllKqnWuNMYd9L0TkVGPMF8EMSIWH9qMX5o4bwvNbCql3tHGo2sp9l0zhP2eOYMGytdz9VjY7inRgqPKPP8OjHgJuACrxLBfe05MBdYavB6tJe7CUUrBSRM4HEJE78HxeKfWNfKMXnrhuJvnlDdQ52sACv140iWtmj8INxEfZuHLWiKNnCD5+3UxdFlQn5NegUWPMXhFxG2MqRCRkjp/QCpZSqp0FwHMicj+wDs/sKqU65FsWfGrDAWIibXznr+upa3Zy57cn8MKWQh5ck8+hmibW5VVoI7vqNH8SrGoRuQmIF5ErCaHtydERVqwW7cFSSgFwMp5dzxvwzOwbDuwLakQqZPmWBUtqHdQ72jhY3UhcpJW3fjyH8WkDeGv7YR29oLrFnwTrv/DsoqkEZnpfhwSLxUJ8VISOaVBKgad9YZExplBETgfeAqYGNyQVqrJMOQerGrFZLRysbmT4oFgq65r528f5OnpBBYQ/uwjtIrIMiPG+NQDPbJiQEBdto1GXCJVSnmnrTgBjzGYRmRPsgFRoab9TMMuUs7+inqS4KJ7+/kzOzkzV0QsqoPzZRfgInsF6JYAFz5bnMzv7oOOdUG+Mye3sfY4VFxVBg1awlOq3RORlY8x/GmOcInKHMebP3t96i06MZ1B9m29JsLjGQZvLRV5ZPTYLvPXjMxmZHK+jF1TA+bNEeBpwkjHG1c1nHe+E+su6eU/iomw0aQVLqf6s/V75RYAvwdLygzoqy5RTUNWI0+niYHUjt54zjn9uKuDap7ay6OShOnpBBZw/CdZePMuDjd181vFOqO+2+KgIGrTJXSnl0T6pcnf4p1S/cOyS4IGKegYPiObFG0/n9LHJtLpclNU6jo5e0GVBFUj+JFgjgYMistf72m2M6fQSIcc/ob7bYqNs1DS2BOJWSqnw5O7gWvVjR5cEax243W5ySuqwWeCNm89gxGDPkuCG/EpdElQ9xp8E66oAPetrJ9SLyFRjjKM7N42PtnG4RitYSvVjk0XkRTzVq/bXk4IblgoGX9Xq3Z3F5JTYSYqNIKekniVzx/DS1iKueVKXBFXv6DDBEpElxpgngR/x9W+Fd3bhWcc7od7Whft8RVxUhE5yV6p/a39u4GMdXKt+oP2RN82tLg7XOCi1W3j82hlcMCUdm9WiS4Kq15yoglXk/dntnX5eXzuh3hjT0N2bxkXZdJK7Uv2YMWZtsGNQoSHLlFNa62C+pPK3T/YSHWFlxKBYImwWXRJUva7DBMsYs9J7+QIwC0/FyQJkdOVBJzihvlvioiJ0krtSSvVT7RvZTWkddY5W/vbJXi6ZnkFNYytfHq7l8bX7WLYqT5cEVa/ypwfrDTwVp2F4lvSKgRU9GVRnxEfZaHG6aHW6iLT5c3a1UkqpvqD9kuDUYUmszC6lzeXm7u9M5AdzxtDc5uL85WvJGBjLRdMydElQ9Sp/EqwkY8zZIvIkcBuwqodj6pRY74HPjS1OkmI1wVKqvxKRYcD9QArwGrDLGLMluFGpnpRlyim1O/jvBeO44+VdDIi20dzm5p+bDlJZ38K6vAoy0xL48xWnaGKlep0/CZavMT3eGNMkIlE9GVBnxUd7/hUaW9pIio0McjRKqSD6O54ho3cD6/CcHHF6UCNSAdd+SXBfRT0DoiK46bkvmDQ0kb9fP5NnPy3QRnYVEvwp+bwpIr8BdorIZqCuh2PqlDhvBUuHjSrV78UYYz7GM6vPAN0aAaNCj29JcPnqPOqbW1m9p4zNB6q5YHIar/3oTFITotmQX8lF0zK4bcF4FkxM0+RKBY0/hz0/7LsWkfeB/B6NqJPiozz/CjqqQal+r1lELgBsInI6mmD1Ob4lwSevn8ktL26nocVJlM3KnmI7D32cr7OtVEg50RysFXQ8Ffnqngmn845WsHRUg1L93Q+BB/CcFPEz4ObghqMCLbvYztRhSVz++KdU1DXz8NWnsutwjS4JqpB0ogpWWAzpi2vXg6WU6teagKeMMatE5FY8w41VH+Dru9pWUM3GvZUMjI3g5R+ewaSMRB7J2quzrVRIOtEcrLUAIpIK/BrIBLKB+3onNP/Et9tFqJTq114CHvdeVwPPE6AzT1Xw+PquTGkdh440gQWa29x8sLuEO9/8UpcEVcjyp8n9ZSAH+CWwH3iuRyPqpKNjGrTJXan+Lt4Y8xqAMeZFIC7I8agAyDLl7C62U3SkiZmjB7HpF+eSPCCK0loHSxdm8vh1M3VJUIUkf8Y0YIzxLRfuFJGAT2PvDl+Tu/ZgKdXvtYjIQmAzcBrgCnI8qot8S4K7D9fy7s5iSmsdXDglneX/eQoxkTYWnZxBXKRNlwVVSPMnwcoVkWuAT4AZQJWIZAIYY/J6Mjh/xEXrEqFSCoAleJrcHwL2ADcFNxzVFb4lwZJaB61OF3srGoi0WXjg8mnERNpodbpYl1fB0oWZwQ5VqRPyJ8Ga4P21pN17j+PZYXhuTwTVGVE2KzarRZvclernjDF7gcXBjkN1T5Ypp7jGQfKAKNbnV3L7gnE8vaGACx9cz6KTh+ooBhU2/EmwrjXGHPa9EJFTjTFf9GBMnWKxWIiLsumgUaX6ORG5E/g50IjnYHq3MaZLh9Or3tV+OntuiZ2q+mZyS+388dKpXHnaSBxtLh3FoMKOPwnWShFZaoz5SETuAK4FpvdwXJ0SHxWhFSyl1BVAhjGmMdiBKP+1P7D51JGD+GhPGW0uN3+9cjoXnZJBq9PFhvxKHcWgwo4/uwgXAD8Tke3ASELsbC+ny40bNzuKaliTU4bT1dFsVKVUH1eAZxaWCiO+6ezLrziFNTnlRNksxEba+L+PDPd/mMvihzfqkqAKS/5UsE4GhgIb8FSuhgP7ejIof/m++dQ2tRITYWP56jxWbC3UbbtK9U9RwJci8qX3tdsYEzKnTqjjyy62M2loIlc9sQW3280rPzqTd3cV65KgCnv+JFj3AIuMMYXe873eAqb2aFR+8n3zmTIsCSsWXrhxNosf3kiWKddSslL9z/3BDkD5z9d39XnBEdbtrSAlPooXbzqDUclx/OJ1XRJU4c+fBGueMcYJYIzZLCJzejgmv2UX25k7PoXKumbW5JYTYbUwLzOFPcV2/YupVP/zJXABEImnyT0DWBvUiNRx+VYf9lU0UFTdAG5oaHHy2udFrM+v1CVB1Sd02IMlIi8DGGOc3uZ2n7d6PCo/Tc5IZH1+BVOGJVHd0EJBZQPr8iqYlJEY7NCUUr3vNeBsPPOvrgfODG44qiNZppy95fUU1zQyKjmerJ/N1+nsqs85UZN7+68Pi9pdh8z/6+dLKumJMfzz0wIArnt6q37zUaofM8b8CDDAQmBQkMNR7ThdbtbklPHQmnyeWL+fwqpGxgwZwMs3ncGoIfEsOjmDk1IGsGBimiZXqk/wZxchfDWpCpltejarhcevm8kvLpiAzWph0tBE/eajVD8mIjFAPJ7PqQFBDkd5+ZYEl6/OI7vYzub91WCBZ26YxZAB0Uens+vqg+pLTtSD5e7gOqTYrBYumJLOycOTqGlq1eRKqf7rYeCnwEdAEZ6dzyoE+DYkLTlrLHe8upNpwxPZV9HAFY9v1unsqs86UYI1WURexFO9an89qVci66Rpwwfy8rYi2pwuImz+FuaUUn2FMeZ137WIvGqMsQczHvVv2cV20hJjWPrKDmaOGszTN8zirx/n6ygG1aedKMG6ot31Yx1ch4xpI5J4ZlMBeyvqmZCuZWal+gsR+Zsx5lYR+ZR21XYRwRjTpUZ3EfkV8F08s7UewbMb8Rnv/XcDtxhjXN2NvS9rf/zNrqIa1uSWM3vMYP5xwywibVadzq76vA4TLGNMWG1vPnn4QAB2FdVqgqVU//I7788bCMAkdxGZj2cH4hwgDvgZsAy4yxiTJSKPARcDb3b3WX1V++NvhsRHk5VXQaTNgr2plb9+vFeXBFW/4M8crLAwJjmehJgIdhyq4YpZI4IdjlKqlxhjyryXTxpjzgrALS/AM1PrTSAR+B/gRv49U+sD4Hw0weqQr+fqP2eO5O63dzN3/BAq65pZOCmNSJtVlwRVv9BnEiyr1cLJw5PYdagm2KEopYKjQUSW4xnT4AIwxvy9C/cZAowCvgOMAd4BrMYY3/JjHZDU/XD7ruxiO0Pio7n77d2cIyk8eu0MHlyTT6TNym0Lxgc7PKV6xTcmWCIyDM8RFCl4BvntMsZs6crDju1rMMY81ZX7dGTa8IH8fd1+HK1OYiJtgby1Uir0bfL+9DX1dHX3cxWQa4xpAYyIOID2ZfEEQL/JHYev7+qT3HK2F9Vw1rhkHr12BjarhXV5FSxdmBnsEJXqNf5UsP4O/Bm4G1gH/BM4vbMP6qCvIaCmjRhIm8tNdrGdGaN0xqBS/Ykx5l4RGcpXj8rpig3A7SKyDM9B9/HAGhGZb4zJAi4EPglAyH2Kr+8qt6SOQzVN2Kywo6iWv6zO0+NvVL/kzzyDGGPMx3hOpjeAo4vPat/X8C7wXhfv06FpvkZ3XSZUqt8RkaeANcA4eh6cAAAgAElEQVR6YBuwvCv3Mca8B2wHtuL5rLoFuAO417tTMQpPNV+1k2XKMWV1HK5pYs64ZD6/ayGD4yP1+BvVb/lTwWoWkQsAm4icTtcTrK/1NYjIhHZ9Dd2WnhRDWmI0O4s0wVKqH5oITAYeB+6kG0mQMebnx3n77K7erz/415clHD7SxIxRg3ji+pnERUWw6OQM4iJtOopB9Uv+VLB+iGf78xA8y3o3d/FZVcBKY0xLu0pYShfv1aGpw5L4dH8VD63JZ01OGU5XyA6hV0oFVp33C1u8MaYST6VJ9SDf+YK/emMXb+8oJtJm5bHrZhAXFaHH36h+z58KVhPwlDFmlYjcChzp4rOO19dQ1cV7HZfT5eZAZQNl9mZqGltZvjqPFVsLtTStVP/wuYj8DCgWkZcA3enSg3w9VwVVjRRVN2LBTXx0JN97eivzMlN01pXq9/ypYL3Ev7ckVwPPd+VBx+trMMY4u3KvjmSZ8qMVq3MmpPDmj+dQUusgy5QH8jFKqRAiIgkAxpg78SwP/gp4Ac+OZdVDskw5RUeaqGtqJSk2ko9+ejbpidEsmJB69Pgb/XKr+jN/KljxxpjXAIwxL4rIkq4+rIO+hoDJLrZz7oRUnt5YwJ5iO3PHpzAvM4U9xXbtAVCq79olImuAvxtjtnrfezeYAfUH2wtrqKpvpqnFyas/OpMxKQM4W1J11pVSXv4kWC0ishDYDJyGd4BfKJqckcjy1XkMjI3gYHXj0R4Anb2iVJ82Hk+16jcikgY8DTxvjKkLblh9j2/O1ZeHa3l/VzGV9S08cf0MJmUk6uetUsfwJ8FaAjwAPATsAW7q0Yi6Yb6ksmJrIfvKG1hryln8cI32ACjVxxlj2oA3gDdEJB34HrBSRLKNMTcGN7q+o/35ghYsHKhqJD7axl9W5/NFYY32XCl1jG9MsIwxe4HFvRBLt9msFh6/bibXPLGZ7BI7v714ip53pVT/cgQ4CFTgGQejAsR3vuB/nDqce97dw/fPGM3WgioWTEjV8wWVOg5/jsq5E/g50IhnOrLbGNPVCck9zma1cNrYZLYUVHPW+CH6l12pfkBE5uGpXM0D3gJ+7h0HowIku9jOmCHx/O79HM6dkMrdF03igY+M9lwp1QF/lgivADKMMY09HUygjE6Ow+2GouomxqUOCHY4SqkeJCL7gb3Ak8CPjDGtQQ6pT/H1Xe06VMOa3HJOGhLHg1eegsvt1p4rpU7AnwSrAM8srLAxKjkegMLqBk2wlOr7zjPG7A92EH2Rr+/qcE0TxUeacLuhpLaZhz/Zq+cLKvUN/EmwooAvReRL72u3MebqHoyp20YnxwFQUBk2RTelVBdpctVzskw5JbUOkgdEk19Wz3M/mMWv39p99HxB7blSqmP+JFj393gUATY4PoqE6AgOVjUEOxSllApb2cV2Im1WNuRX8odLpzI3M1XPF1TKT/5Mcv8SyMBzUPNo4MyeDCgQLBYLI5PjKKjSCpZS/ZWI+PMFUp1AfXMbO4pquOa0kVx12kg9X1CpTvDnA+g1IA+YiueA5rDIWkYnx7OnxB7sMJRSwfMe8K1gBxFufE3tn5gKXtp6kKTYCLYX1XD/h7k660qpTvDrG54x5kci8jSeoaPrejakwBiVHMfK7FLanC4ibP4U6pRSfYkxRpOrTjra1H6kiUM1TYCFqcOS+N6Zo8ktqdO+K6U6wa8ES0RigHjADYTFtrzRyfG0udwU1zgY6W16V0r1XSKSCvwayASygfuMMUeCG1V48TW1Dx4Qxb6KBlbcOJt73t2D1WLRWVdKdZI/pZ2HgZ8CHwFFQG6PRhQgvqSqQBvdleovXsbz+fRLYD/wXHDDCT/ZxXbiomxs3FvFvRdPZtaYZOZlprCnWNstlOosf47Ked13LSKvGmPC4m/aaO8srIPVYdEyppQKAGPMo97LnSJyRVCDCUMWYFvBERafksGVs0boAc5KdUOHCZaI/M0Yc6uIfIpnadD3PsaYkN9JmJoQTUyklYOVWsFSqp/IFZFrgE+AGUCViGQCGGPyghpZCPM1tW85UM2Lmw8SF2XDlNXxp5VGm9qV6oYTVbB+5/15A2E2yR3AarUwanC8jmpQqv+Y4P31X3iKMQCP4/mCeG6wggplvqb2kloH1Q0tNLQ4OW3MIJbMHatN7Up1U4cJljGmzHv5pDHmrF6KJ6BGJcdxQCtYSvULxphzRCQZOAnYb4ypDHZMoS7LlFNqdzA/M4WHs/bxwOUn84+NBdrUrlQA+LOLsEFElgMGcAEYY/7eo1EFyKjkOLLyKnC53Fj1G5hSfZqIXA78HsgBpojIPcaY54McVkjLLraTmZrAo2v3cempw/iPGSPYV9HAnmK7TmpXqpv82UW4CagB0oChQHqPRhRAo5LjaWlzUVbnCHYoSqmetxSYYYxZDEwHbg9yPCHvpJR43ttVTHpiDPd8d7JOalcqgPzZRXiviAwFIvH0NWT0eFQB4ttJWFDZyNCk2CBHo5TqYS5jTD2AMaZORPSbVQd8je2PZu2jxekmKsLKo1n7tKldqQD6xgRLRJ4CzsAzaDQWz3yZ03s4roAY5Z2FdbCqgTNOSg5yNEqpHrZPRP6M57SJecC+IMcTknyN7fll9RysbiQ5PpKk2EhiImza1K5UAPmzRDgRmAysBCbhOY8wLGQMjCXSZtGdhEr1D0vwfAFc6P15Y3DDCU1ZppxDR5qob25l4tBE1v78HNpcbqYMS2TBxDRNrpQKEH+a3OuMMW4RiTfGVIpIVI9HFUCD46NYk1PGrNGD9JuZUn3be8aY84MdRKjbfbiW5jYndQ4nL9w4jQHRkUentWtju1KB408F63MR+RlQLCIvAbYejikgfGXwhmYn1Q0tLF+dx03PfYbT5f7mf1gpFY5qROS7IjJBRDJ9Q0bVV9U2tXGgspGfLhzPhPREbWxXqoecaJJ7gjGmzhhzp4gk4Bk2eiGwtdei6wbffJdLpw/j9S8O8cbNZ3LJI5vIMuX6LU2pvikFz7mpPjpg1MvX1P7pvipe2FzAwNhI3ttVgt3Rpo3tSvWQEy0R7hKRNcDfjTG+pOrdXogpILKL7cwdn0JGUgwNLU4Kq5u0DK5UHyUiicAiY4w2XB7DV80vrXVQ3dhCq8vNjKEJ3HDWGJ3WrlQPOtES4XjgX8BvRGSbiNzsrWSFhckZiazPr2DBxFQirBZe2HxQy+BK9UEiciuwE88BzxcEO55Q46vmXzQtg+IaB/ctnkqto+3otHZtbFeqZ3SYYBlj2owxbxhjvgNcBCQCK0Xkie48UERSRaRIRCZ05z7fZL6kkp4Yw43Pfs6o5Dj++WkBqQnRWgZXqu+5GhA842R+EuRYQk52sZ2Thw1k2ao8zpuYypWnjThazVdK9Rx/mtwBjgAHgQpgTFcfJiKReA5f7fHDo21WC49fN5OlCzM5deQgXG741pR0/aamVN/jMMa0eM8eDKtdzr1h4tAE3tpxmEibhfsumUqby63VfKV6wQnHNIjIPOB7eIb2vQX83BhjuvG8B4DHgF914x5+s1ktLJiYxrkTUtleVMOLW4v4z1kje+PRSqng0G9Qxyi1O2hscZIxMIZnNhVoU7tSveREuwj3A3uBJ4EfGWNau/MgEfk+UGGMWSkivZJg+VgsFq6ZPZJ7393D7sO1TBmW1JuPV0r1rMki8iKe5Mp3DYAx5urghRU8vl2Dm/dX8dynBzl9zCCWzB1Ljja1K9VrTlTBOs8Ysz+Az/oB4BaR84BTgGdF5LvGmNIAPqNDl546nPs/yOX+D3OZNXowkzMS9UNGqb7hinbXjwUtihDRftdgnaON5jYXVquVcyakcd6k9GCHp1S/0WGCFeDkCmPMPN+1iGThqYr1SnIFMCA6gsEDotiQX0lmWgLLV+exYmshj183U5MspcKYMWZtT9xXRFKBz/EcvdMGPINnttZu4BZjjKsnnttdvl2DS+aO5Scv7+BXF07gnZ3FOgNQqV7mb5N72Msy5cRG2nADwwbG8uaP51Bc08SDq/N4aE0+a3LKdMq7Ugo47oacZcBdxpi5eJYiLw5WbN8ku9jO7DHJ3PevHE4ensSSuWN116BSQdDpBEtE/Dm/8ISMMfONMbndvU9nZBfbOX9yOjNHDeL37+/h129+id3RxkvbimhqdepROkqp9nwbcoq9r2cAvkrZB8B5wQjKH5MzEnlr+2Eq6pr5/eIpuNy6a1CpYOhKBeu9gEfRC3yDRx+99lRumDOG1z4/xKEjTcwaM4ifnpfJmz+eQ0mtgyxTHuxQlVJdICLDROR5EVkpIjeKyOwu3uf7eDfktHvbYozxffuqA0J2p8yQAdFUNbQwOD6KD3aXsvjhjbprUKkg6HSCZYz5Vk8E0tN8g0e//49tREVYGRgXRVyUjfd3lfLE+v1E2qxaRlcqvP0deBrPLKx1wINdvM8PgIXeXtFTgGeB9tlJAlDT9TB7htPlZtWeUm567nMSom385juTiIu0sXRhpvaaKhUE35hgeSevPygiH4jIAyIyqDcCC7T2g0fjIm1cO3skY4bEM2v0IF79rIiWNqeW0ZUKbzHGmI8Bt3den6MrNzHGzDPGnG2MmQ/sAK4HPhCR+d4/ciGwPgDxBoxv5+D/vpNNqd1BUlwU7+0q5sfnjNOjcJQKEn8qWC8DucAvgf3Acz0aUQ/yDR69bcF4bj8vk6FJMRTXOCioauSCv6zXMrpS4a3ZexahTUROp4sJVgfuAO4VkU/xVMheC+C9uy3LlHO4pol6Rxunjx3Mx3ecrS0PSgWZXw3rxphHvZc7ReSKE/7hMOGraK3MLuEnL+1kaFKMltGVCm8/xNOcPgT4GXBzd2/orWL5nN3d+/WU7GI70RE26prbuOe7k4mKsB1tedDRDEoFhz8JVq6IXAN8gmcnTZWIZAIYY/J6MrieZrNa+PbUDNaaSt7dVYyj1Ul8dLc3SSqlguMy4GZjzJFgB9LbhibFsLOohsXThzEhPZFWp4t1eRUsXZgZ7NCU6rf8WSKcACwBngeWAoPxzIfpMxOTr5g1nMYWJ+9/WRLsUJRSXRcJrBKRF9r1S/ULWw5UAZBdXMv9H+bqzkGlQsA3lmuMMeeISDJwErDfe2J9n3LqyEGMTYnn1c+KuGLmiGCHo5TqAmPMA8ADIjIL+B8RecIYMz7YcfW0/LI63vjiMN+fM5qzxg1hT7FdzxtUKgR8Y4IlIpcDvwdygCkico8x5vkej6wXWSwWLjt1OP+30nDvO9mcNX6IfjgpFWZEJBbPMuH38Exb/01wI+pZvgOd/9+/coi2Wfnx/HGkJERrz5VSIcKfJcKlwAxjzGJgOnB7z4bU+5wuN5v3e0rsOw7V6FR3pcLTLmAynj6s84wxK4IdUE/xjWW47/0c9lU0kBAbya/e2KWfWUqFEH8SLJcxph7AGFNHYLc+h4QsU86RxhbmSwrFNU28ctMZusVZqTDR7viu6cD/AoUiEiUiUUEMq0f5DnROS4pmyIBoPvrpPP3MUirE+JNg7RORP4vIxSLyZ2BfTwfV27KL7cwdn8L3zhhNmb2Z1z8/pFPdlQofz3p/folnZp/x/urV8057U3axnQnpiXy6r5ob545hYFyUfmYpFWL8SbCW4BkwutD788YejSgIfOcUzhmXzOwxg1m+Ko+s3HKd6q5UGDDGXO29vMIYM9YYM8YYMwbPkTd90uSMRFbtKSM+ysZVs0ceHcugn1lKhQ5/hj69Z4w5v8cjCaL5ksqKrYVc8sgmJqQnsOVANYmxkbrFWakwICJn4em9+qmILPO+bQVuBaYELbAeNC51ALVNrSTHR/Fo1j7W5VXoWAalQow/CVaNiHwXyANcEP4DRo/lm+qeZcrZU2xn9pjBbC88QpndQcbA2GCHp5Q6sRogHYgGhnrfcwE/D1pEPezZTw9is8AvL5xAaa1DxzIoFYL8SbBSgJ+2e+0Gzu2ZcILHd07hgolpLJ4+jHP/nMXPXt3J6WOTmZyRqB9eSoUoY8xuYLd37lWx730RiQxiWD2mtqmVl7YWctG0DC7XuX1KhawTJlgikggsMsY09lI8ISFjYCwZSbFs2lfFsIGxfLSnlBVbC/WsQqVC20UicgeezzUL0Ar0mbNifHOvntlUQEOLkxvmjAl2SEqpE+iwyV1EbgV24jng+YLeCyn4skw5cdE2kmIj+KLwCH+67GSKa5p4cHUeD63JZ01Omc6bUSr03IjnQOYPgBuAPcENJ3B8c6+Wrcrj84NHiI+28deP8/VzSKkQdqJdhFcDApwB/KR3wgkN2cV25ksqj107k9qmVi55ZCNl9mZe2lZEU6tTB5EqFZoqjTElQIIxJgvPual9gm/u1ffOGE1ji5MHrzxF514pFeJOlGA5jDEt3rMH++zAvuPxjW2YOXoQ//rvuYxOjqeqoYXRQ+K45ZxxvPnjOfrhplToqRWRxYBbRG7C0z/aJ2QX25k7bgjPbCpgfOoAFkxI07lXSoU4f+Zggaefod+YL6mkJ8aw+OGN/GNTAdWNrQyMjWTbgSNc9NcNZB+2M3JwHE9tOMCanDJa2lysySnT5UOlgmsJcBD4JZ7eq5uDG07gTM5I5MPsMvaU2PnBWWNoc7l17pVSIe5ETe6TReRFPMmV7xr4ymC/PunYsQ1XzRrBmtxyHr5mOktf3sklj2zEZrWwcFIay1YZfvH6LtISY5iXmcLy1XnaEK9ULxKRY+f0pQAr6UOV9/mSyi9e34XNAvsr6ln88Eade6VUiDtRgnVFu+vHejqQUNN+bIPT5WZPiZ3/969cpg5L4mNTTpvLTbndQeqAGExpPRdOTufq00aydGEmix/eSJYp11PtleodV3Xwvhv4qDcD6SmHjjRSWd/CoqlDSYyJ1LlXSoWBDhMsY8za3gwklLWvaD214QAXTE7n9LGDeW5zIbuL7bS53Dy3pZC3dhbz1PdmHe2N0ARLqZ5njLkh2DH0tH9sLCDSZuE3F00iLTEm2OEopfzgz6BRxb8rWgDLV+fx0OzpfO/MMazMLuX2Fdv5+beE5zcXcu2Tm0lLjOGS6cN4aE2+DilVqpeISAmeqpUFzw7C/caYicGNqnucLjf/+rKEF7YcZMaoQQwZEB3skJRSfvK3yV15tW+Av//DXB5cnceAmAhe/+Iw8zJTsFig6EgTr39xWEc6KNWLjDFDjTEZxpiheJrcNwc7pu7wzb667/0cWp1uKuqa9bNEqTDSawmWiESKyHMisl5EtnrPNww7vuXCpQsziYu0ccf5wqZfLmDpwkyS46P4/pwxxEfZOFzTRGVdM3+7ajp5ZXXc8coO3WGoVC8xxhwEJgQ7ju7IMuWU1DqwWeG00YP58CfzdDyMUmGkN5cIrwWqjDHXiUgysB14pxefHzDtG+B9fK8fWpPPNaePBCz8Y8MBXv/iENERNgqrG1m+Oo8XtxzkytNGklNSp8uHSgWQiKzAs0QIkAGUBTGcbssutjMqOY7sYjt3fnsSkTar9ncqFUZ6M8F6FXit3eu2Xnx2r5mckcjy1Xm8+eM5jB0Szx8+yKG2qY2cYjsXTElnZXYZ+eUNLDp5qCZcSgVW+93ODuCzYAUSCJMzEnly/X4GxUWycFIarU4X6/IqWLqwzxyvqFSf1msJljGmHkBEEvAkWnf11rN703xJZcXWQhY/vJG4KBuRNiunjRnM8IGxfLC7hKZWF6W1TZTWOlgyZyx3vf0le8sb+LY34dIZWkp1WSFwEeDbZnc28KfghdM9U4clUedoY3B8FMtX57Eur0JnXykVRnp1F6GIjADeBB4xxrz4TX8+HLUf6fDuzmLK7c08+4PTiIm0kTEwlmc/LWDKsCSyTDlvbj8MwOB4NwsnpekMLaW6523gDeBIsAMJhHd2FuMGbj9vPLWNrTr7Sqkw02sJloik4Rn6d6sxZk1vPTcYfD1a8yWVm577jMse3cS8zBTe3+X5wHz6+7OIsFr45Rtf8v6uYuxNbfzHo5v44byxDB8Uy1MbDgDoh6lSnVNkjLkn2EEEgtvt5pXPijhlxECuP2N0sMNRSnVBb1aw7gQGAXeLyN3e9y40xjT1Ygy96tgjd+5eNImXthUdTbi2HqgiwmZl1dJ53P9hLo+t3Y8FOHNcsvZnKdV574rIH4E9vjeMMc8GMZ4u21FUQ15ZPX+4dGqwQ1FKdVFv9mDdDtzeW88LFcfuODx3YtrXEq7v/2MbIwfHEWmzgBs27a3ikukZrMwu+0p/liZcSp3QlUAO4BsuGrYzUV75rIjYSBvfOXlosENRSnWRTnLvZR0lXE9tOMDCiWn89uIpPLp2H//cVECby81JsRFcdupw/vvccZx235qjCdeyVYaH1uRzzoRUpg5LYu74FNbnV5BdbNfkS/VXzcaYm4MdRHc4XW5WZpfw+ueHmTl6EHFR+hGtVLjSv71BduwRPElxkdz9nUmAmxe8Zx2et2wtibERNDtdjEmJ59QRA9l6oJrdh2tJS4xh1Z5SfvH6LtISY5iXmaLVLtVfHRSRXwFf4K1eGWPC5rBn3+T2nJI6WpwuSmod3PTcZ7qrWKkwpQlWiGg/3mFeZgqr9pQRGWHlX7fPZfP+av756QHySuvJMhVkmYqj/9xHe8oYFBtBraONaSMGMiE9gStnjuA7f92g1S7V30TiOSLHNyjKjWdjTVjIMuWU2h2MTI4lKsLKyp/M5ZJHNumuYqXClCZYIaKjhvjbVmxnXmYKTS1OBsRE8M4tc3hi/QGONLaw61Ats8cMJqfUjqOygbWmgjU5nmM0rBYYOTiWgbGRWC0WckrrSE04frVLZ2+pvsAYc0OwY+iO7GI7s8ck8+ynBfxgzhiiImw6uV2pMKYJVgjxpyH+1hXbGTk4jk9MOWeelMz/XT6N1Tll3P7Sdv569XRGDIrj/g9z2bK/ioLqRv7wQe7R+6/KKSMtMZrqhhYWTkzj8hnDWbowk4v/toEHV+cRYbMyMT0BLOjSogo7IlKCp2plAQYD+40xE0/8T4WOyRmJ3PvuHlqdbs6WFJ3crlSY0wQrhHWUcO0+XEtxTRNl9mYe+Miw1pQzIDqCv328l3mZKRysaiA60sa2X5/HQ2vyKat1sLWgmhmjBrHrUC1VlmZWbCtixbYiJmckUFLr4MWthVw2YzhLX90JwDWzR2ovlworxpijW+5EZBRwT/Ci6bz5kspv3s7GaoEsU8F97+fo5HalwpgmWGGkfcJ167njj1a37jhfjvZVHTtvq321689XnHK02nXvdyfR1Ork+U8LONLYCsAr24pwuyEmworVYuGxa2Zw4YPrOxwVodUuFaqMMQdFZEKw4+gMqwVcLhfThg8kITpCJ7crFeY0wQpTx1a3gE5Vu17cWsi8zBQaWl0MGxjD9WeM5s3thymzO6hqaOGRrH08krUPqwWGD4ohPsrGf8wYzh8/yGV3sZ1vTx2q1S4VUkRkBf+efZUBlAUxnE7LL6+nxN7MbQsyuXr2yGCHo5TqJk2w+ih/q11XzRrBmtxyfnDWGMalDmDZqjzSE2NYPH0Y6/Mr2HqgmsM1Dh74KO/ovR2tzTyzsYDoSCuxkTacLjd3fXsiS579jPzyBhYdZ+eiJluqFzzW7toBfNaVm4hIJPA0MBqIBn6PZzr8M3gSuN3ALcYYVzdi/Zos49mgMl9SAnlbpVSQaILVD5yo2uV0udlTYmfxwxs5a/wQCqsaATjS2EJhdSPRkTa2/vo8LBb4vw8Nr31+iO+dOZot+6soPNJIub2ZJ9Yf4In1nvMTm1ub+PJQDc1tbvZW15ExMJbVOWVfq27pqAgVSCLyQ+BpY0ybiMwFphtjtnTxdtcCVcaY60QkGdgO7ADuMsZkichjwMV4Dq4PmCxTQWbaADIGxgbytkqpINEEq587djzEsiumgQVyS+q+dnbiR3tKcbrd3Dz/JE4ensSyVXkkxURy9eyRbD1Qzce55cwaMxhTWkdJrQOADXsrOX3MYNbmVbLrUC3nT05n5e6Sr4yK+KY5XZqMqRMRkXuAKcDzQBtQBPxURFKNMb/twi1fBV5r97oNmAGs9b7+ADifACZY9c1tbCuo5gdzxgTqlkqpINMESx23wrVwUjpw/FERlz266SvVrsM1Tew8VIPVauGxa2fw93X7Ka1t4hNTQcbAGHYeqqXF6aKivoUXthQSHWGlxeliUkYiCTERuFywp8zOgOgIVmaXfi356kwypslXv3QhcLoxxje9vUBE/hPYBHQ6wTLG1AOISAKeROsu4AHf/YE6ICkQgfts2lt5dDyDUqpv0ARLndCJZnN1VO1qv3Pxietn8dCafJ7ffJCbzh5LfHQEr35WRH55PZv3V7Mur/LoszbuqwI8Q4xGJsdhs1iYODSJvLLDXD5jOBdPyyC3xM6n+6sYNjD2G48I0l2O/UZ9u+QHAGNMq4jUdfWGIjICT4XqEWPMiyLyp3a/nQDUdPXe7TldbrJMOX/9ZC/REVamjxgUiNsqpUKAJliqU/ypdh27c/H9XcW0OF1cf8ZoYiJtDBkQ7RmMetUp7D5sp7Kumc0HqvnW5HS2Fx0hp9jO3vJ6dhTW4PL+Z/PRtft5dO1+rBZITYim1O5A0hLJKyvm8pkjOG9iGpeeMoxLHt3I3vIGvjU1/Rt3OWr1q89oEpGxxpj9vjdEZCz/3lHYKSKShueInVuNMWu8b28XkfnGmCw8FbNPuhnz0bMHS2sdFFQ1EhVh5bYVX+ipCkr1ERa3u0ufQQEnIqOBA2vWrGH48OHBDkd1k++b+Z5iOxPSE3hpWxGldgfzMlNYa8opr2smLTHma9Uu35yuB688hYUT03l752F++fqX3LEwk88LazhU3cj+ygZanS7aXMf//26EFZL/f3t3Hl9VdS1w/HdvRkMISMIQCggVsgQU7BOUMj8BLfih2PoqPq1aW9Fq64cqfe918rX1tc9nB7TUoeJQLVX4WEWtdWQQmZRBGYSGFUAkhBBCSEgYQtTk44AAABMCSURBVIabvD/2SbyGJICeey+5Wd/Px4+59yZn7XNyslhnn332Tk/h8PFaEoMBxgzIYuS5Wfz6tVy6dEhm6pBslm8/wP6Kqsber+V5xSQGg/bUYxQVFBQwYcIEgH6q+vFn3Y6IDAbmA0uAj4A+wOXAjaq64TNs7w/AdGBb2NszgTlAMpALzFDVUNjP9OU089eS3P3cvziP3141hMlzVvI/085nwbp87pqUY0vjGNMGnCyHWQ+WiYjWbi2GTxXR0jxdc5bsYOOect7RYjqmJvLypkL6dEljZ8kRRp6bydzrh/Haln38x982M2uSMKBHOs++t5tVOw8y8txM8g8eI0CAI1W1vL6liFc/LALgWHUlf3l3N53Tkig9Ws2X+nSme0YKwUCQ3KLDdO2YetKnHpveerSesNhS1a3ek4PTcPNffQDco6qf6Rahqs7EFVRNjfvsrTzR1sIKxgzoyrsflQJw6cBuFJZX2tqDxsQJ68EyMRfe2zUorGBp+nrL3nKWbiumtq7+hJ6wsTldeXVzIWXHalj304ms2lHC7EV51NfX84OJOazeWcLCD/ZyxZBsUhITeCfvAIWHKqmtqyfUpCesX+ZZFBw6TsZZSUwc2J1N+WWUHK2me0Yqowdk8eyafMDdelzRpCdsxfYDdO+Y0uI4MCvGPuFXD9aZ4PP0YPXsdBa5RRUsnTWeKx9aZT1YxrQRJ8thVmCZNqW1Yiz8VuToAVk8+55XCI3ow2sf7mssvlKTEnhzaxEzF2zg/ukXsjH/EMWHj7P+4zKG9+3C9uLDbNlbccIAnk6pCXRITaLsaDWBAAzK7kR6SiIrd5QwfVgvRg/oSnJigDvmb6TzWUlMvqAHz60vcG04hWKsvRVc7b3ACtXVc8u89SzTA/Tvmk5iQoDsTqk2BsuYNsIKLNOuNB371fCU4+mMA5uzZDvPrNnN7eP7k5mezMIP9rKtqIKs9BSqauo4XhOiOlRHTaiOY9UhqmpbntA7MRggOTHAsHO6kJgQ4J28Er5yfndG9Mukf9cOzJj3PpkdUpgyJPuEcWCtzQcWD09ItvcCC0CLKrj8gRVcPqg7Vw/v3SZ/j8a0VzYGy7QrpzqnV2vjwF7dXEhVbR3XXtKH1KQEUpMSmLlgA/dMG0xiMNh463HWZUJtXT0z52/gF18dzNDenXly5S7e3FrE9OG92VpYQcmRKg4dq2H97rLG25Gvbi7i1c1Fje1LDFbzUfERjlaF2HvoCB1SEnllUyGlRzeRlZ7CyP5ZzH7rk/nARg/Isick48S6j8sA+PGUgfTN6hDj1hhj/GQFlmk3WlsyKHy9xqYz2IcPvA+fYHXdx6UszztAemoi897bze7SY6zZdZB6YNZl0jgOLCs9mXu/foErxhZs4IHpFzKkV2fufS2XJduK6dc1nY0Fh9hfUQXAam8+MICyYzVsLz5CQhDq6iCzQzJbCspJTUogKRggLTmBH16Ww21/3cCWwgrG5XTl9Q9Pb6Z8K75iZ+2uUrpnpHBOZlqsm2KM8ZkVWMZw6k89Np1gtelnLc1231iMpSTyx6U7GJvTlU0Fh0gIBpg/YwRzl39EeWUNy7SYyRdks7vkKB/uLWdQdgaXfDGTVzYXsqP4CBXHayk4VMnRKjdDwOxF2xv3obIm1DjmC6BLWhKV1SEgwLYity7kySZntYIreurr61m7q5SL+2USCNjxNibeWIFlTDNa6+1q0HDrselnLc1231oxFj4O7M6JOSzO3c+i3P38eMp5XD44mx6dUpm5YAP3XXUBicEgv3tLqamt46qLerPjwGEWb93PjLFf5JzMDixYm0/uvsMcrgrxzJrd1ITcOMvFufvp3jG1cXqKfpkdmDSwGzc8uY4dxUeZMiT7hN4uK7YiZ09pJUUVx7m4X5dYN8UYEwFWYBnjs9bGgUHzxdipzAcWfpuyoLQSgPLKatbtKqUOuHnMF0lNSiA5McjMBRu496oLyS08TMGhY6zeeZCBPTLYVXKUpIQgS7cdYNE/ixvbdFZSkIrKGurrP+ntOt35wKwYOz1rdrlbwZdYgWVMXLICy5gYCi/GwseBNe3tau02ZWtjxsJ7xubeMKxxpvwHr/0S5/XI4DdvbGOZFpOZnsLz7xc0PhG5YnsJU4dk89LGQrYXH+UKr3ertYH289fm2xQDp2HNrlLOTkuif9f0WDfFGBMBVmAZc4Y4lduSn/cJyYbi60FvHNjWwnISE4K89L1R/Omdnewpdb1dPTql8sIHe6kO1bGvvJIPdpeREAxQerSagdkZ7Cw+QnJCkABQXVvPwttG8rWHV7NMi22SzFPkxl91IWgFqTFxyQosY+LAqT4hearjwB67YTizFynz3t3NkF6dqA7Vs+9QFYnBAOs/LiNUV0+oro5QPTy5ahdnpyUxNqerLfNyivaVV5JfeowbR/aNdVOMMREStQJLRILAw8BQoAq4WVV3RCu+Me3VyXrGWhoH9urmQuqBx28c/qnZ7x+4Zuin5gPrlJbE7xfl0bNTKv92US/mLNluY7JaEaqr58mVuwAIeK/tOBkTf4JRjHUlkKqqXwZ+BPw+irGNMS1oKMBmTsxh4e2juGtSDmlJCdx9xSAu7tuFqx5ZzX1vbOMPi/Max3at/biU/IPH2FNayfk9O5GWHKSw/DgvbyqksibE/YvzuHXe+hPWeWzvQnX13DpvPS+8X0BSQoAXPiiw42RMnIpmgTUaeANAVd8DhkUxtjHmFDQUW3dMGMCkwT2Ye8OwxoJr1mXC6h9N4K5JOaQnJzL76qHMnj6UjNQkrh/Rl5TEICWHq5g6pCcv3j6KfeXHWabFJw/ajizTYooqjnN2h2RG9c/ipe/ZcTImXkWzwMoAysNeh0TExoAZcwYLL7gmDOxOcmLwUwXYpEE9uGPCADqkJHL18N50Tkvm5U17SUoINo7JMp/YWljBmP5ZlB2rYcLA7nacjIlj0SywKoCO4bFVtTaK8Y0xETK4ZwYb8st4686x3DUph5pQHcvzDjCoZ0asm3ZGGdwzgxU7Slg6axzfvKSPHSdj4lg0e5BWAVOB50RkBPBhFGMbYyJovHRj/tp8rn70XcbmdGV53gGyO6UyXrrFumlnlIbjdN3ja+w4GRPnollgvQhMEpHVuIdnbopibGNMBCUEAzx6/bDG6SDumpRjTxE2w46TMe1H1AosVa0DvhuteMaY6GpuOghzIjtOxrQP0RyDZYwxxhjTLliBZYwxxhjjMyuwjDHGGGN8ZgWWMcYYY4zPzqSJPhMAioqKYt0OY0wUhP2tJ8SyHT6x/GVMO3OyHHYmFVjZANddd12s22GMia5sYGesG/E5Wf4ypv1qNoedSQXWOmAMsA8IxbgtxpjIS8AlpnWxbogPLH8Z0/60msMC9fW2irsxxhhjjJ9skLsxxhhjjM/OpFuEp0REgsDDwFCgCrhZVXdEOGYS8DTQF9f9P0NVt0Uw3iXAfao6XkS6AY8BZ+O6I29QVV/Hq3j79yRu/1KAX6nq373PrgXuUNUv+xwzAbdfgjumN+EWA/+j97oKt6/7/Yzrxe4GvA9MAmqBp4B6YAvwPW/VAd81iZsK/MmLn4c7j32NKyIbgHLv5S7g117MZNzxvUZVD/oc88fAV70YD6vqE977ETmP2ppY5C8vbtzmsFjkL2/b7SqHRTt/eTHbdA5riz1YVwKp3k7+CPh9FGJOARJVdSRwD+6XHBEi8p/A47gTGOA3wDOqOhb4GXBeBMJ+EzioqmOAycCDXlsuBL6DWzvSb1MBVHUU8N/AbOAPuBN4PLAQ+C+/g3rJ+FGg0ntrNvAzb98DwDS/Y7YQ9+fAPao6GvePwhU+x0sFUNXx3n83AXNx+zoWl6RyfI45HhgJjALGAb299yN5HrU1schfEN85LBb5C9pRDot2/vJitvkc1hYLrNHAGwCq+h4wLAox84BE7+ozA6iJYKydwNfDXo8CeonIYuA6YFkEYv4NuDvsda2IZAL/B/wgAvFQ1ZeAW7yX5wD7cVcjG733EoHjEQj9O9wfZqH3+iLgHe/r14GJEYjZXNwNQBcRCeCuev0+p4YCaSLylogsFZEvA92AqSKyDBgBrPU55uXAh7iF3V8B/hHp86gNikX+gvjOYVHPX9Ducli08xfEQQ5riwVWBp90GQKERCTStzqP4Lqft+G6hOdEKpCqvsCnT9a+QJmqTgTyicAVkaoeUdXDItIReB6XrJ4A7gQO+x0vLG6tiDyN61J/XlX3AYjISOD7wP1+xhORbwEHVPXNsLcDqtrwpMdhoJOfMVuJux13HuUC3fH/H51juKR4OW6R9WeBwcBi4F+BLsCNPsfMwhUM3/BizsfduonoedTGxCJ/QRznsFjlLy923OewGOUviIMc1hYLrApcxdwgqKq1EY55J/CmqubgquqnG7ovo+Ag8Hfv61eI0BWviPQG3gbm4f54BgCPAAuAQSLyQCTiquqNuG7ex0Skg4hMx10pXaGqB3wO921gknf1cyHwF9wVUYOOwCGfY7YU92lgjKqe5732+1ZRHvBXVa1X1TygBEBV3/aS8T/w/1w6iPs7qVZVBXrhfrcRP4/akFjkL4jzHBar/AXtIofFIn9BHOSwNjfIHViFu/f9nIiMwHXnRVoZn1yRlQJJRG/26ZW48RPzgLHAVr8DiEh34C3g+6q6xHt7sPdZX2CBqvra1S4i1wO9VPVe3JVKHfA14FZgvKqW+hkPwLtv3xB/Ge4K5bciMl5Vl+HGb7wdpbgv4f6xBdftPsrnsN8GLgBuF5GeuMT7voiMUdUVROZcWgnMFJHZuLlh9gLnq2ooUudRGxSL/AVxnMNikb+8bbeLHBaj/AVxkMPaYoH1Iq6aXo0bcHZTFGLeDzwpIitwTxb8RFWPRiEuwCzgcRG5DXdr4doIxPgJ7gmfu0WkYSzDZFWtbOVnPq+FwJ9FZDku2f8A+DPuFsJCEQF4R1V/HsE2gDu+j4lIMq67+/kIx2twM7BARGqBamCGz9t/AnhKRFbini76NnAUeMi7JbUL/2/V/ENExuLGRQRxTzPZpJufFov8BfGdw2KRv6B957BI5y+IgxxmE40aY4wxxvisLY7BMsYYY4w5o1mBZYwxxhjjMyuwjDHGGGN8ZgWWMcYYY4zPrMAyxhhjjPFZW5ymwbTAW0fpOeCfuMdaz8KtAfbHJt/3FaCPqs49jW1/CyhtWET1FH+mL27ekBFN3j8bN0PvANxcPHuAW1W1/ISNnKFE5Bbgz6oaySVHjGk3LH9Fj+Wv6LACK/4sVdVrAEQkBVARmaeqjbP7quobp7tRVX3KvyYyH3hUVV8EEJE7cQuJXuNjjEj7CW4GY0tQxvjH8ld0WP6KAiuw4ltHIIRb/HQZcAA3Id983NXXn7yv9wDnAmtV9TYR6QY8BXTGTYZ4A26R1iLcWmY/xc1a3AOYq6oPicg43ArrAGnez1Q3bZCInAP0aEhOnjlAuvf5dbgJ+6pwS17c4sWeiruizcatWD8NOB/4oaq+LCIfAWu8/diCmwgvA/ir9/9E3CrsS0VkM25x1CG4K+VpqlouIvfiZgcOArNV9W/ecdvoxcrArVE10dv3BcCVJ/81GGM+A8tflr/aNBuDFX8uFZFlIrIUeAa4Q1WPeJ896y24Gj4zbQ7wHeBiYIqI9MAloL+r6kjv64ubxPgC8FXcauZ3egltMPBNVb0Ut+7YN1poX0/cDLyNVDXkJYhM4JfApao6Grem1q3et3VU1SnAfcBtwNdxyathJuxewN2qejEu2V0J/AxY5C318A3gCREJ4hLNfFUdh1sKYbKITAb6qeoo3EKiPxWRzt6213rHbRHw76r6BC5Zt6UrVmPaAstflr/ihvVgxZ/GLvZmaDPv7VDVwwAisg9IBQS3gjiqutT77BdhP7NaVau897fgrrr2AnNE5Aguga1qoQ35uGTSSESScAlkO7C1oT3AcuAy3JXdBu+9Q0CuqtaLSJnXXoB8Vd3R0D5vHwbikjSquldEKoCu3vc0bG+Pt40+wEXeFR+4pS/OaeZ7e7SwX8aYz8/yl+WvuGE9WO1LXTPvNbdWUi4wHEBExorIfU0+v1BEEkQkDXfltx14HLhJVb+FW/wz0FwDVHUvUCIi08Lenom7YtuFW628g/f+ONyK6i21M9wXvKtXcAuPbvX2Y4y3H1/A3V442ML2tgFvq+p44FLcYNuPWoldh/39GBNNlr8sf7UpdoBNc/4XmOZdDf0SN4AzXBLwOrAC+JWqlgDzgDUisgo3dqJnK9u/HrhWRFaIyBrgX4AZ3nZ+DrwtIu8BWcAjp9jmKuBBb3uFwCveflzqLcb6EnCLqta28POvAEe8xXDfB+rDrkSbswJ4TUSaTcTGmJix/GX564xgiz2b0+I9Sv3dVrrxY0JEilTVur+NMS2y/GWiyXqwjDHGGGN8Zj1YxhhjjDE+sx4sY4wxxhifWYFljDHGGOMzK7CMMcYYY3xmBZYxxhhjjM+swDLGGGOM8ZkVWMYYY4wxPvt/OxFAhdWaXgcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 720x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10,4))\n",
    "\n",
    "ax1.plot(range(0, pca.n_samples_), pca.explained_variance_ratio_*100, 'o-', markersize=5, markerfacecolor=\"None\", markeredgewidth=1)\n",
    "ax1.set_ylabel('Prop. Variance Explained')\n",
    "\n",
    "ax2.plot(range(0, pca.n_samples_), np.cumsum(pca.explained_variance_ratio_)*100, 'o-', markersize=5, markerfacecolor=\"None\", markeredgewidth=1)\n",
    "ax2.set_ylabel('Cumulative Prop. Variance Explained')\n",
    "\n",
    "for ax in (ax1, ax2):\n",
    "    #ax.set_ylim(bottom=-0.05, top=1.05)\n",
    "    ax.get_xaxis().set_major_locator(mpl.ticker.MaxNLocator(integer=True))\n",
    "    ax.set_xlabel('Principal Component')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 10.6.2 Clustering the Observations of the NCI60 Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "agg_complete = AgglomerativeClustering(affinity='euclidean', linkage='complete').fit(StandardScaler().fit_transform(df_ncl60))\n",
    "agg_average = AgglomerativeClustering(affinity='euclidean', linkage='average').fit(StandardScaler().fit_transform(df_ncl60))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, (ax1, ax2) = plt.subplots(2, 1, figsize=(15,10))\n",
    "\n",
    "color_threshold = 48\n",
    "\n",
    "ax1.set_title('Complete linkage')\n",
    "ax2.set_title('Average linkage')\n",
    "\n",
    "plot_dendrogram(agg_complete, color_threshold=color_threshold, ax=ax1, labels=df_ncl60.index)\n",
    "plot_dendrogram(agg_average, color_threshold=color_threshold, ax=ax2, labels=df_ncl60.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "agg_complete = AgglomerativeClustering(affinity='euclidean', linkage='complete', n_clusters=4).fit(StandardScaler().fit_transform(df_ncl60))\n",
    "fig, ax = plt.subplots(1, 1, figsize=(10,6))\n",
    "plot_dendrogram(agg_complete, color_threshold=60, ax=ax, labels=df_ncl60.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### K-means"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Kmeans:\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1\n",
      " 1 1 1 1 0 3 3 3 3 3 3 3 3 3 3 0 0 0 2 2 2 2 2 2 2 2 2]\n",
      "Hierarchical:\n",
      " [0 0 0 0 3 3 3 3 0 0 0 0 0 0 0 0 0 3 3 3 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1\n",
      " 1 1 1 1 0 2 0 2 2 2 2 2 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0]\n"
     ]
    }
   ],
   "source": [
    "kmeans = KMeans(n_clusters=4, n_init=20).fit(df_ncl60)\n",
    "print('Kmeans:\\n', kmeans.labels_)\n",
    "print('Hierarchical:\\n', agg_complete.labels_)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>Hierarchical</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>K-Means</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Hierarchical   0  1  2  3\n",
       "K-Means                  \n",
       "0             30  0  0  7\n",
       "1              0  8  0  0\n",
       "2              9  0  0  0\n",
       "3              1  0  9  0"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(kmeans.labels_, agg_complete.labels_, rownames = ['K-Means'], colnames = ['Hierarchical'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Use PCA first"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = make_pipeline(StandardScaler(), PCA(n_components=5), AgglomerativeClustering(affinity='euclidean', linkage='complete', n_clusters=4))\n",
    "model.fit(df_ncl60)\n",
    "fig, ax = plt.subplots(1, 1, figsize=(10,6))\n",
    "plot_dendrogram(model.named_steps['agglomerativeclustering'], color_threshold=60, ax=ax, labels=df_ncl60.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}